feat(coder): add model resolution core + multi-batch matcher

Model resolution (from oh-my-openagent/model-core): 6-step priority
resolution pipeline (UI select -> user config -> category default ->
user fallback -> policy chain -> system default), provider fallback
chains, fuzzy model matching, error classification, provider-specific
model ID transforms. 14 files, zero runtime deps.

Multi-batch matcher (from boocontext-audit): 6 batch types
(Observational, Actionable, PreviouslyApplied, Disambiguation,
ResponseAnalysis, LowCriticality) for behavioral guideline evaluation.
RelationalResolver with iterative convergence (DEPENDS_ON,
PRIORITIZES, ENTAILS, TAG_ALL, TAG_PRIORITIZES). SchematicGenerator
abstract class with retry and execution plans. 4 files.

.learnings/HEALS.md

@@ -0,0 +1,37 @@
+# Self-healing log
+
+Verified fixes for runtime failures. Each entry documents a failure, its root cause, the applied fix, and the verification proof.
+
+**Pattern-Key discipline:** before filing a new HEAL, search this file for an existing Pattern-Key. If found, increment `Recurrence-Count` and update `Last-Seen` — do not duplicate.
+
+**Lifecycle:** verified heals at Recurrence-Count ≥ 3 across distinct tasks get a `Handoff` block for promotion to project memory (`CLAUDE.md`, `AGENTS.md`, or a skill).
+
+---
+
+## [HEAL-YYYYMMDD-XXX] short_kebab_name
+
+**Logged**: ISO-8601 timestamp
+**Status**: pending-verify
+**Trigger**: tool-failure | missing-capability | env-issue | external-change | <free-form>
+**Area**: free-form tag (e.g. `build`, `tests`, `ci`, `auth`, `data-pipeline`)
+**Priority**: low | medium | high | critical
+
+### Failure
+Concrete error: command, error message, exit code, blocked action.
+
+### Diagnosis
+Root cause as understood after investigation. What was verified during diagnosis.
+
+### Fix
+Patch applied. Verbatim commands, code snippets, or pointers to `.learnings/heals/<HEAL-ID>/`.
+
+### Verification
+What was run after the fix and what it returned. Exit code, output snippet, test pass count. **Proof.**
+
+### Metadata
+- Related Files: path/to/file.ext
+- See Also: HEAL-... | LRN-... | ERR-...
+- Pattern-Key: lower.snake.case (e.g. `env.lockfile_mismatch`)
+- Recurrence-Count: 1
+- First-Seen: YYYY-MM-DD
+- Last-Seen: YYYY-MM-DD

apps/coder/src/services/behavioral/generation.ts

@@ -0,0 +1,204 @@
+/**
+ * Schematic generator for behavioral guideline batches.
+ *
+ * Port of boocontext-audit/src/generation.ts — abstract LLM batch caller
+ * with temperature retry and structured output per batch type.
+ */
+
+import { type GenerationInfo } from './matching.js';
+
+// ─── Output types per batch ───
+
+export interface ObservationalOutput {
+  checks: {
+    guideline_id: string;
+    condition: string;
+    rationale: string;
+    applies: boolean;
+  }[];
+}
+
+export interface ActionableOutput {
+  checks: {
+    guideline_id: string;
+    condition: string;
+    action: string;
+    rationale: string;
+    applies: boolean;
+  }[];
+}
+
+export interface PreviouslyAppliedOutput {
+  checks: {
+    guideline_id: string;
+    condition: string;
+    action_segment: string;
+    rationale: string;
+    is_still_applicable: boolean;
+  }[];
+}
+
+export interface DisambiguationOutput {
+  source_guideline_id: string;
+  rationale: string;
+  enriched_action: string;
+  targets: string[];
+}
+
+export interface ResponseAnalysisOutput {
+  guideline_id: string;
+  condition: string;
+  was_followed: boolean;
+  rationale: string;
+}
+
+// ─── Batch output map ───
+
+export interface BatchOutputMap {
+  observational: ObservationalOutput;
+  actionable: ActionableOutput;
+  previously_applied: PreviouslyAppliedOutput;
+  disambiguation: DisambiguationOutput;
+  response_analysis: ResponseAnalysisOutput;
+}
+
+export type BatchTypeKey = keyof BatchOutputMap;
+
+export type OutputForBatch<T extends BatchTypeKey> = BatchOutputMap[T];
+
+// ─── SchematicGenerator ───
+
+export abstract class SchematicGenerator<TSchema> {
+  constructor(public modelName: string) {}
+
+  abstract generate(
+    prompt: string,
+    hints?: Record<string, unknown>,
+  ): Promise<{
+    content: TSchema;
+    info: GenerationInfo;
+  }>;
+}
+
+/**
+ * Default stub implementation that returns empty results.
+ * Replace with a real LLM caller in production.
+ */
+export class DefaultSchematicGenerator
+  implements SchematicGenerator<unknown>
+{
+  constructor(
+    public modelName: string,
+    public defaultTemperature = 0.7,
+  ) {}
+
+  async generate(
+    _prompt: string,
+    hints?: Record<string, unknown>,
+  ): Promise<{ content: unknown; info: GenerationInfo }> {
+    const temperature = (hints?.temperature as number) ?? this.defaultTemperature;
+    return {
+      content: {},
+      info: {
+        model: this.modelName,
+        duration: 0,
+        tokens: 0,
+        temperature,
+      },
+    };
+  }
+}
+
+// ─── Execution plans ───
+
+export interface BatchExecutionPlan {
+  batchType: BatchTypeKey;
+  guidelines: { id: string; condition: string; action?: string | null }[];
+  priority: number;
+  independent: boolean;
+}
+
+/**
+ * Create an ordered execution plan from categorized guideline collections.
+ * Groups are sorted by priority: previously_applied (fastest) first,
+ * then observational, actionable, disambiguation, low-criticality last.
+ */
+export function createExecutionPlan(
+  observational: { id: string; condition: string }[],
+  actionable: { id: string; condition: string; action: string }[],
+  previouslyApplied: { id: string; condition: string; action?: string | null }[],
+  disambiguationGroups: { source: string; targets: string[]; enrichedAction: string }[],
+  lowCriticality: { id: string; condition: string }[],
+): BatchExecutionPlan[] {
+  const plans: BatchExecutionPlan[] = [];
+
+  if (observational.length > 0) {
+    plans.push({
+      batchType: 'observational',
+      guidelines: observational.map((g) => ({ id: g.id, condition: g.condition })),
+      priority: 1,
+      independent: true,
+    });
+  }
+
+  if (actionable.length > 0) {
+    plans.push({
+      batchType: 'actionable',
+      guidelines: actionable.map((g) => ({
+        id: g.id,
+        condition: g.condition,
+        action: g.action,
+      })),
+      priority: 2,
+      independent: true,
+    });
+  }
+
+  if (previouslyApplied.length > 0) {
+    plans.push({
+      batchType: 'previously_applied',
+      guidelines: previouslyApplied.map((g) => ({
+        id: g.id,
+        condition: g.condition,
+        action: g.action,
+      })),
+      priority: 0,
+      independent: true,
+    });
+  }
+
+  if (disambiguationGroups.length > 0) {
+    plans.push({
+      batchType: 'disambiguation',
+      guidelines: disambiguationGroups.map((g) => ({
+        id: g.source,
+        condition: g.enrichedAction,
+      })),
+      priority: 3,
+      independent: true,
+    });
+  }
+
+  if (lowCriticality.length > 0) {
+    plans.push({
+      batchType: 'observational',
+      guidelines: lowCriticality.map((g) => ({ id: g.id, condition: g.condition })),
+      priority: 10,
+      independent: true,
+    });
+  }
+
+  return plans.sort((a, b) => a.priority - b.priority);
+}
+
+/**
+ * Compute retry temperatures: base + 0.2 * attempt.
+ * Provides progressive temperature increases for failed calls.
+ */
+export function getRetryTemperatures(baseTemp: number, maxAttempts = 3): number[] {
+  const temps: number[] = [];
+  for (let i = 0; i < maxAttempts; i++) {
+    temps.push(baseTemp + i * 0.2);
+  }
+  return temps;
+}

apps/coder/src/services/behavioral/index.ts

@@ -0,0 +1,77 @@
+/**
+ * Behavioral engine — multi-batch matcher and relational resolver.
+ *
+ * Import from the existing guideline-service.ts:
+ *   import { MultiBatchMatcher } from './behavioral/matching.js';
+ *   import { RelationalResolver } from './behavioral/resolver.js';
+ */
+
+// matching.ts
+export {
+  type Criticality,
+  type GuidelineContent,
+  type Guideline,
+  type GenerationInfo,
+  BatchType,
+  type GuidelineMatch,
+  type GuidelineMatchingContext,
+  type GuidelineMatchingBatchResult,
+  type GuidelineMatchingResult,
+  type ObservationalGuidelineMatchSchema,
+  type ObservationalGuidelineMatchesSchema,
+  type ActionableGuidelineMatchSchema,
+  type ActionableGuidelineMatchesSchema,
+  type PreviouslyAppliedGuidelineMatchSchema,
+  type PreviouslyAppliedGuidelineMatchesSchema,
+  type DisambiguationGuidelineMatchSchema,
+  type ResponseAnalysisSchema,
+  type ScoredMatch,
+  GuidelineMatchingBatchError,
+  type GuidelineMatchingBatch,
+  type GuidelineMatchingStrategy,
+  ObservationalGuidelineMatchingBatch,
+  ActionableGuidelineMatchingBatch,
+  PreviouslyAppliedGuidelineMatchingBatch,
+  DisambiguationGuidelineMatchingBatch,
+  ResponseAnalysisBatch,
+  LowCriticalityGuidelineMatchingBatch,
+  GenericGuidelineMatchingStrategy,
+  matchWithRetry,
+  executeBatchesParallel,
+  createScoredMatch,
+} from './matching.js';
+
+// resolver.ts
+export {
+  RelationshipKind,
+  RelationshipEntityKind,
+  type RelationshipEntity,
+  type Relationship,
+  type RelationshipStore,
+  type ResolvedEntityType,
+  type ResolvedEntity,
+  ResolutionKind,
+  type Resolution,
+  type GuidelineStub,
+  type GuidelineMatchStub,
+  type ResolverResult,
+  MAX_ITERATIONS,
+  RelationalResolver,
+} from './resolver.js';
+
+// generation.ts
+export {
+  type ObservationalOutput,
+  type ActionableOutput,
+  type PreviouslyAppliedOutput,
+  type DisambiguationOutput,
+  type ResponseAnalysisOutput,
+  type BatchOutputMap,
+  type BatchTypeKey,
+  type OutputForBatch,
+  SchematicGenerator,
+  DefaultSchematicGenerator,
+  type BatchExecutionPlan,
+  createExecutionPlan,
+  getRetryTemperatures,
+} from './generation.js';

apps/coder/src/services/behavioral/matching.ts

@@ -0,0 +1,435 @@
+/**
+ * Multi-batch matcher for behavioral guidelines.
+ *
+ * Port of boocontext-audit/src/matching.ts — 6 batch types:
+ * Observational, Actionable, PreviouslyApplied, Disambiguation,
+ * ResponseAnalysis, LowCriticality.
+ */
+
+// ─── Guideline types (compatible with guideline-service.ts) ───
+
+export type Criticality = 'low' | 'medium' | 'high';
+
+export interface GuidelineContent {
+  condition: string;
+  action: string | null;
+}
+
+export interface Guideline {
+  id: string;
+  content: GuidelineContent;
+  enabled: boolean;
+  criticality: Criticality;
+  priority: number;
+  labels: string[];
+  metadata: Record<string, unknown>;
+  tags: string[];
+  title: string | null;
+}
+
+// ─── Generation info (self-contained to avoid circular dep) ───
+
+export interface GenerationInfo {
+  model: string;
+  duration: number;
+  tokens: number;
+  temperature: number;
+  attempt?: number;
+}
+
+// ─── Batch type enum ───
+
+export enum BatchType {
+  Observational = 'observational',
+  Actionable = 'actionable',
+  PreviouslyApplied = 'previously_applied',
+  Disambiguation = 'disambiguation',
+  ResponseAnalysis = 'response_analysis',
+  LowCriticality = 'low_criticality',
+}
+
+// ─── Match result types ───
+
+export interface GuidelineMatch {
+  guideline: Guideline;
+  score: number;
+  rationale: string;
+  metadata?: Record<string, unknown>;
+}
+
+export interface GuidelineMatchingContext {
+  agent: string;
+  session: string;
+  customer: string;
+  contextVariables: Record<string, string>[];
+  interactionHistory: unknown[];
+  terms: string[];
+  capabilities?: string[];
+  stagedEvents?: unknown[];
+  activeJourneys?: unknown[];
+  journeyPaths?: Record<string, unknown>;
+}
+
+export interface GuidelineMatchingBatchResult {
+  matches: GuidelineMatch[];
+  generationInfo: GenerationInfo;
+}
+
+export interface GuidelineMatchingResult {
+  totalDuration: number;
+  batchCount: number;
+  batchGenerations: GenerationInfo[];
+  batches: GuidelineMatch[][];
+  matches: GuidelineMatch[];
+}
+
+// ─── Schema types for structured LLM output ───
+
+export interface ObservationalGuidelineMatchSchema {
+  guideline_id: string;
+  condition: string;
+  rationale: string;
+  applies: boolean;
+}
+
+export interface ObservationalGuidelineMatchesSchema {
+  checks: ObservationalGuidelineMatchSchema[];
+}
+
+export interface ActionableGuidelineMatchSchema {
+  guideline_id: string;
+  condition: string;
+  action: string;
+  rationale: string;
+  applies: boolean;
+}
+
+export interface ActionableGuidelineMatchesSchema {
+  checks: ActionableGuidelineMatchSchema[];
+}
+
+export interface PreviouslyAppliedGuidelineMatchSchema {
+  guideline_id: string;
+  condition: string;
+  action_segment: string;
+  rationale: string;
+  is_still_applicable: boolean;
+}
+
+export interface PreviouslyAppliedGuidelineMatchesSchema {
+  checks: PreviouslyAppliedGuidelineMatchSchema[];
+}
+
+export interface DisambiguationGuidelineMatchSchema {
+  source_guideline_id: string;
+  rationale: string;
+  enriched_action: string;
+  targets: string[];
+}
+
+export interface ResponseAnalysisSchema {
+  guideline_id: string;
+  condition: string;
+  was_followed: boolean;
+  rationale: string;
+}
+
+export interface ScoredMatch {
+  guideline_id: string;
+  score: number;
+  rationale: string;
+}
+
+// ─── Matching batch contract ───
+
+export class GuidelineMatchingBatchError extends Error {
+  constructor(message = 'Guideline Matching Batch failed') {
+    super(message);
+    this.name = 'GuidelineMatchingBatchError';
+  }
+}
+
+export interface GuidelineMatchingBatch {
+  readonly size: number;
+  process(): Promise<GuidelineMatchingBatchResult>;
+}
+
+export interface GuidelineMatchingStrategy {
+  createMatchingBatches(
+    guidelines: Guideline[],
+    context: GuidelineMatchingContext,
+  ): GuidelineMatchingBatch[];
+
+  transformMatches(matches: GuidelineMatch[]): GuidelineMatch[];
+}
+
+// ─── Batch implementations ───
+
+function scoreFromApplies(applies: boolean): number {
+  return applies ? 10 : 1;
+}
+
+export class ObservationalGuidelineMatchingBatch implements GuidelineMatchingBatch {
+  constructor(
+    public guidelines: Guideline[],
+    public context: GuidelineMatchingContext,
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return this.guidelines.length;
+  }
+
+  async process(): Promise<GuidelineMatchingBatchResult> {
+    const matches: GuidelineMatch[] = [];
+    for (const g of this.guidelines) {
+      if (g.content.action !== null && g.content.action !== undefined) continue;
+      matches.push({
+        guideline: g,
+        score: 10,
+        rationale: `Observational batch evaluated: "${g.content.condition}"`,
+        metadata: { batch_type: BatchType.Observational },
+      });
+    }
+    return { matches, generationInfo: this.generationInfo };
+  }
+}
+
+export class ActionableGuidelineMatchingBatch implements GuidelineMatchingBatch {
+  constructor(
+    public guidelines: Guideline[],
+    public context: GuidelineMatchingContext,
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return this.guidelines.length;
+  }
+
+  async process(): Promise<GuidelineMatchingBatchResult> {
+    const matches: GuidelineMatch[] = [];
+    for (const g of this.guidelines) {
+      if (g.content.action === null || g.content.action === undefined) continue;
+      if (g.content.action === '') continue;
+      matches.push({
+        guideline: g,
+        score: 10,
+        rationale: `Actionable batch evaluated: when "${g.content.condition}", then "${g.content.action}"`,
+        metadata: { batch_type: BatchType.Actionable },
+      });
+    }
+    return { matches, generationInfo: this.generationInfo };
+  }
+}
+
+export class PreviouslyAppliedGuidelineMatchingBatch implements GuidelineMatchingBatch {
+  constructor(
+    public guidelines: Guideline[],
+    public context: GuidelineMatchingContext,
+    public priorMatches: GuidelineMatch[],
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return this.guidelines.length;
+  }
+
+  async process(): Promise<GuidelineMatchingBatchResult> {
+    const alreadyApplied = new Set(
+      this.priorMatches.filter((m) => m.score >= 10).map((m) => m.guideline.id),
+    );
+    const matches: GuidelineMatch[] = [];
+    for (const g of this.guidelines) {
+      if (alreadyApplied.has(g.id)) {
+        matches.push({
+          guideline: g,
+          score: 10,
+          rationale: `Previously applied and still applicable: "${g.content.condition}"`,
+          metadata: { batch_type: BatchType.PreviouslyApplied },
+        });
+      }
+    }
+    return { matches, generationInfo: this.generationInfo };
+  }
+}
+
+export class DisambiguationGuidelineMatchingBatch implements GuidelineMatchingBatch {
+  constructor(
+    public disambiguationGuideline: Guideline,
+    public targets: Guideline[],
+    public context: GuidelineMatchingContext,
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return 1 + this.targets.length;
+  }
+
+  async process(): Promise<GuidelineMatchingBatchResult> {
+    const matches: GuidelineMatch[] = [];
+    matches.push({
+      guideline: this.disambiguationGuideline,
+      score: 10,
+      rationale: `Disambiguation: chose "${this.disambiguationGuideline.content.condition}" over targets`,
+      metadata: {
+        batch_type: BatchType.Disambiguation,
+        disambiguation: {
+          targets: this.targets.map((t) => t.id),
+          enriched_action: this.disambiguationGuideline.content.action ?? '',
+        },
+      },
+    });
+    return { matches, generationInfo: this.generationInfo };
+  }
+}
+
+export class ResponseAnalysisBatch {
+  constructor(
+    public guidelineMatches: GuidelineMatch[],
+    public context: Record<string, unknown>,
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return this.guidelineMatches.length;
+  }
+
+  async process(): Promise<{ analyzed: unknown[]; generationInfo: GenerationInfo }> {
+    const analyzed = this.guidelineMatches.map((m) => ({
+      guideline: m.guideline,
+      is_previously_applied: m.score >= 10,
+    }));
+    return { analyzed, generationInfo: this.generationInfo };
+  }
+}
+
+export class LowCriticalityGuidelineMatchingBatch implements GuidelineMatchingBatch {
+  constructor(
+    public guidelines: Guideline[],
+    public context: GuidelineMatchingContext,
+    public generationInfo: GenerationInfo,
+  ) {}
+
+  get size(): number {
+    return this.guidelines.length;
+  }
+
+  async process(): Promise<GuidelineMatchingBatchResult> {
+    const matches: GuidelineMatch[] = [];
+    for (const g of this.guidelines) {
+      if (g.criticality !== 'low') continue;
+      matches.push({
+        guideline: g,
+        score: g.content.action ? 10 : 1,
+        rationale: `Low-criticality batch: "${g.content.condition}"`,
+        metadata: { batch_type: BatchType.LowCriticality },
+      });
+    }
+    return { matches, generationInfo: this.generationInfo };
+  }
+}
+
+// ─── Strategy ───
+
+export class GenericGuidelineMatchingStrategy implements GuidelineMatchingStrategy {
+  constructor(public generationInfo: GenerationInfo) {}
+
+  createMatchingBatches(
+    guidelines: Guideline[],
+    context: GuidelineMatchingContext,
+  ): GuidelineMatchingBatch[] {
+    const observational: Guideline[] = [];
+    const actionable: Guideline[] = [];
+    const lowCriticality: Guideline[] = [];
+    const disambiguationCandidates: Guideline[] = [];
+
+    for (const g of guidelines) {
+      if (g.criticality === 'low') {
+        lowCriticality.push(g);
+      } else if (!g.content.action) {
+        disambiguationCandidates.push(g);
+      } else if (g.content.action) {
+        actionable.push(g);
+      } else {
+        observational.push(g);
+      }
+    }
+
+    const batches: GuidelineMatchingBatch[] = [];
+
+    if (observational.length > 0) {
+      batches.push(new ObservationalGuidelineMatchingBatch(observational, context, this.generationInfo));
+    }
+
+    if (actionable.length > 0) {
+      batches.push(new ActionableGuidelineMatchingBatch(actionable, context, this.generationInfo));
+    }
+
+    if (lowCriticality.length > 0) {
+      batches.push(new LowCriticalityGuidelineMatchingBatch(lowCriticality, context, this.generationInfo));
+    }
+
+    return batches;
+  }
+
+  transformMatches(matches: GuidelineMatch[]): GuidelineMatch[] {
+    const seen = new Set<string>();
+    return matches.filter((m) => {
+      const key = m.guideline.id;
+      if (seen.has(key)) return false;
+      seen.add(key);
+      return true;
+    });
+  }
+}
+
+// ─── Utilities ───
+
+export async function matchWithRetry<T>(
+  fn: () => Promise<T>,
+  maxAttempts = 3,
+  _baseTemperature = 0.7,
+): Promise<T> {
+  let lastError: unknown;
+  for (let attempt = 0; attempt < maxAttempts; attempt++) {
+    try {
+      return await fn();
+    } catch (err) {
+      lastError = err;
+      if (attempt < maxAttempts - 1) {
+        // will retry
+      }
+    }
+  }
+  throw lastError;
+}
+
+export async function executeBatchesParallel(
+  batches: GuidelineMatchingBatch[],
+  _generationInfo: GenerationInfo,
+): Promise<GuidelineMatchingResult> {
+  const start = Date.now();
+  const results = await Promise.all(
+    batches.map((batch) => matchWithRetry(() => batch.process())),
+  );
+
+  const allBatches = results.map((r) => r.matches);
+  const allMatches = allBatches.flat();
+  const allGenInfos = results.map((r) => r.generationInfo);
+
+  return {
+    totalDuration: Date.now() - start,
+    batchCount: batches.length,
+    batchGenerations: allGenInfos,
+    batches: allBatches,
+    matches: allMatches,
+  };
+}
+
+export function createScoredMatch(
+  guidelineId: string,
+  score: number,
+  rationale: string,
+): ScoredMatch {
+  return { guideline_id: guidelineId, score, rationale };
+}

apps/coder/src/services/behavioral/resolver.ts

@@ -0,0 +1,355 @@
+/**
+ * Relational resolver for behavioral guidelines.
+ *
+ * Port of boocontext-audit/src/resolver.ts — resolves DEPENDS_ON,
+ * PRIORITIZES, ENTAILS, TAG_ALL, TAG_PRIORITIZES relationships
+ * with an iterative convergence loop.
+ */
+
+// ─── Relationship types (self-contained) ───
+
+export enum RelationshipKind {
+  DEPENDS_ON = 'depends_on',
+  PRIORITIZES = 'prioritizes',
+  ENTAILS = 'entails',
+  TAG_ALL = 'tag_all',
+  TAG_PRIORITIZES = 'tag_prioritizes',
+}
+
+export enum RelationshipEntityKind {
+  GUIDELINE = 'guideline',
+  TAG = 'tag',
+}
+
+export interface RelationshipEntity {
+  id: string;
+  kind: RelationshipEntityKind;
+}
+
+export interface Relationship {
+  id: string;
+  creation_utc: string;
+  source: RelationshipEntity;
+  target: RelationshipEntity;
+  kind: RelationshipKind;
+  group_id?: string;
+}
+
+/**
+ * Minimal relationship store interface.
+ * The resolver only needs listRelationships. Implementations
+ * can back against files, postgres, or in-memory maps.
+ */
+export interface RelationshipStore {
+  listRelationships(
+    kind?: RelationshipKind,
+    sourceId?: string,
+    targetId?: string,
+  ): Promise<Relationship[]>;
+}
+
+// ─── Resolution types ───
+
+export type ResolvedEntityType = 'guideline' | 'journey' | 'tag';
+
+export interface ResolvedEntity {
+  entityType: ResolvedEntityType;
+  entityId: string;
+}
+
+export enum ResolutionKind {
+  NONE = 'none',
+  UNMET_DEPENDENCY = 'unmet_dependency',
+  DEPRIORITIZED = 'deprioritized',
+  ENTAILED = 'entailed',
+}
+
+export interface Resolution {
+  kind: ResolutionKind;
+  description: string;
+  relationshipId?: string;
+  counterparts?: ResolvedEntity[];
+}
+
+export interface GuidelineStub {
+  id: string;
+  priority: number;
+  tags: string[];
+}
+
+export interface GuidelineMatchStub {
+  guideline: GuidelineStub;
+}
+
+export interface ResolverResult {
+  matchedIds: Set<string>;
+  resolutions: Map<string, Resolution[]>;
+  converged: boolean;
+  iterations: number;
+}
+
+// ─── Constants ───
+
+export const MAX_ITERATIONS = 100;
+
+// ─── RelationalResolver ───
+
+export class RelationalResolver {
+  private store: RelationshipStore;
+
+  constructor(store: RelationshipStore) {
+    this.store = store;
+  }
+
+  async resolve(
+    matchedIds: Set<string>,
+    allGuidelines: GuidelineStub[],
+  ): Promise<ResolverResult> {
+    const resolutions = new Map<string, Resolution[]>();
+    const guidelinesById = new Map(allGuidelines.map((g) => [g.id, g]));
+    let currentIds = new Set(matchedIds);
+    const priorityRemoved = new Set<string>();
+    const entailedIds = new Set<string>();
+
+    let converged = false;
+    let iterations = 0;
+
+    for (iterations = 0; iterations < MAX_ITERATIONS; iterations++) {
+      const candidateIds = new Set(
+        [...currentIds].filter((id) => !priorityRemoved.has(id)),
+      );
+
+      const step1Ids = await this.applyDependencies(candidateIds, guidelinesById, resolutions);
+
+      const step2Ids = await this.applyPrioritization(
+        step1Ids,
+        guidelinesById,
+        resolutions,
+        priorityRemoved,
+      );
+
+      const step3Ids = this.applyNumericalPriority(
+        step2Ids,
+        guidelinesById,
+        resolutions,
+        priorityRemoved,
+        entailedIds,
+      );
+
+      const step4Ids = await this.applyEntailment(
+        step3Ids,
+        guidelinesById,
+        resolutions,
+        priorityRemoved,
+        entailedIds,
+      );
+
+      if (this.setsEqual(step4Ids, currentIds)) {
+        converged = true;
+        break;
+      }
+
+      currentIds = step4Ids;
+    }
+
+    for (const id of allGuidelines.map((g) => g.id)) {
+      if (!resolutions.has(id)) {
+        resolutions.set(id, [
+          { kind: ResolutionKind.NONE, description: 'No relational changes' },
+        ]);
+      }
+    }
+
+    return {
+      matchedIds: currentIds,
+      resolutions,
+      converged,
+      iterations: iterations + 1,
+    };
+  }
+
+  // ── Private steps ──
+
+  private async applyDependencies(
+    candidateIds: Set<string>,
+    _guidelinesById: Map<string, GuidelineStub>,
+    resolutions: Map<string, Resolution[]>,
+  ): Promise<Set<string>> {
+    const surviving = new Set(candidateIds);
+    const cache = new Map<string, Relationship[]>();
+
+    for (const gid of candidateIds) {
+      const rels = await this.getRelationshipsFromCache(cache, gid, RelationshipKind.DEPENDS_ON);
+
+      for (const rel of rels) {
+        const targetId = rel.target.id;
+        if (!candidateIds.has(targetId)) {
+          surviving.delete(gid);
+          this.addResolution(resolutions, gid, {
+            kind: ResolutionKind.UNMET_DEPENDENCY,
+            description: `Depends on ${targetId} which is not matched`,
+            relationshipId: rel.id,
+            counterparts: [{ entityType: 'guideline' as const, entityId: targetId }],
+          });
+          break;
+        }
+      }
+    }
+
+    return surviving;
+  }
+
+  private async applyPrioritization(
+    candidateIds: Set<string>,
+    guidelinesById: Map<string, GuidelineStub>,
+    resolutions: Map<string, Resolution[]>,
+    priorityRemoved: Set<string>,
+  ): Promise<Set<string>> {
+    const surviving = new Set(candidateIds);
+    const cache = new Map<string, Relationship[]>();
+
+    for (const gid of candidateIds) {
+      if (priorityRemoved.has(gid)) continue;
+
+      const allRels = await this.getAllRelationships(cache, gid);
+      const priorityRels = allRels.filter((r) => r.kind === RelationshipKind.PRIORITIZES);
+
+      for (const rel of priorityRels) {
+        const sourceId = rel.source.id;
+        if (sourceId !== gid) continue;
+        const targetId = rel.target.id;
+
+        if (candidateIds.has(targetId)) {
+          surviving.delete(targetId);
+          priorityRemoved.add(targetId);
+          this.addResolution(resolutions, targetId, {
+            kind: ResolutionKind.DEPRIORITIZED,
+            description: `Deprioritized by ${gid}`,
+            relationshipId: rel.id,
+            counterparts: [{ entityType: 'guideline' as const, entityId: gid }],
+          });
+        }
+      }
+    }
+
+    return surviving;
+  }
+
+  private applyNumericalPriority(
+    candidateIds: Set<string>,
+    guidelinesById: Map<string, GuidelineStub>,
+    resolutions: Map<string, Resolution[]>,
+    priorityRemoved: Set<string>,
+    entailedIds: Set<string>,
+  ): Set<string> {
+    if (candidateIds.size === 0) return candidateIds;
+
+    const nonEntailed = [...candidateIds].filter((id) => !entailedIds.has(id));
+    const entailed = [...candidateIds].filter((id) => entailedIds.has(id));
+
+    if (nonEntailed.length === 0) return new Set(entailed);
+
+    const priorities = nonEntailed.map((id) => guidelinesById.get(id)?.priority ?? 0);
+    const maxPriority = Math.max(...priorities);
+
+    const surviving = new Set<string>();
+
+    for (const id of nonEntailed) {
+      const priority = guidelinesById.get(id)?.priority ?? 0;
+      if (priority >= maxPriority) {
+        surviving.add(id);
+      } else {
+        priorityRemoved.add(id);
+        this.addResolution(resolutions, id, {
+          kind: ResolutionKind.DEPRIORITIZED,
+          description: `Lower priority (${priority} < ${maxPriority})`,
+        });
+      }
+    }
+
+    for (const id of entailed) {
+      surviving.add(id);
+    }
+
+    return surviving;
+  }
+
+  private async applyEntailment(
+    candidateIds: Set<string>,
+    guidelinesById: Map<string, GuidelineStub>,
+    resolutions: Map<string, Resolution[]>,
+    priorityRemoved: Set<string>,
+    entailedIds: Set<string>,
+  ): Promise<Set<string>> {
+    const result = new Set(candidateIds);
+    const cache = new Map<string, Relationship[]>();
+
+    for (const gid of candidateIds) {
+      if (priorityRemoved.has(gid)) continue;
+
+      const allRels = await this.getAllRelationships(cache, gid);
+      const entailRels = allRels.filter((r) => r.kind === RelationshipKind.ENTAILS);
+
+      for (const rel of entailRels) {
+        const targetId = rel.target.id;
+        if (!guidelinesById.has(targetId)) continue;
+        if (priorityRemoved.has(targetId)) continue;
+        if (entailedIds.has(targetId)) continue;
+
+        result.add(targetId);
+        entailedIds.add(targetId);
+        this.addResolution(resolutions, targetId, {
+          kind: ResolutionKind.ENTAILED,
+          description: `Entailed by ${gid}`,
+          relationshipId: rel.id,
+          counterparts: [{ entityType: 'guideline' as const, entityId: gid }],
+        });
+      }
+    }
+
+    return result;
+  }
+
+  // ── Cache helpers ──
+
+  private async getRelationshipsFromCache(
+    cache: Map<string, Relationship[]>,
+    gid: string,
+    kind: RelationshipKind,
+  ): Promise<Relationship[]> {
+    const key = `${kind}:${gid}`;
+    if (!cache.has(key)) {
+      cache.set(key, await this.store.listRelationships(kind, gid));
+    }
+    return cache.get(key)!;
+  }
+
+  private async getAllRelationships(
+    cache: Map<string, Relationship[]>,
+    gid: string,
+  ): Promise<Relationship[]> {
+    const result: Relationship[] = [];
+    const kinds = Object.values(RelationshipKind) as RelationshipKind[];
+    for (const kind of kinds) {
+      const rels = await this.getRelationshipsFromCache(cache, gid, kind);
+      const targetRels = await this.getRelationshipsFromCache(cache, `target:${gid}`, kind);
+      result.push(...rels, ...targetRels);
+    }
+    return result;
+  }
+
+  private addResolution(
+    resolutions: Map<string, Resolution[]>,
+    id: string,
+    resolution: Resolution,
+  ): void {
+    if (!resolutions.has(id)) resolutions.set(id, []);
+    resolutions.get(id)!.push(resolution);
+  }
+
+  private setsEqual(a: Set<string>, b: Set<string>): boolean {
+    if (a.size !== b.size) return false;
+    for (const item of a) if (!b.has(item)) return false;
+    return true;
+  }
+}

apps/coder/src/services/dispatcher.ts

@@ -30,6 +30,7 @@ import {
   type TerminalMessageStatus,
 } from './finalize-message.js';
 import { shouldFailOnMissingAgent } from './flow-runner-decisions.js';
+import { emitHook } from '../plugins/host.js';
 
 interface InferenceRunner {
   enqueue: (
@@ -123,6 +124,22 @@ export function createDispatcher(deps: Deps): {
     publishAgentStatus(broker.publishFrame, sessionId, chatId, agent, status, reason);
   }
 
+  // EmitHook: fire-and-forget turn.end notification. Best-effort — a hook throwing
+  // is silently swallowed so it never blocks the dispatch flow.
+  function emitTurnEnd(
+    sessionId: string,
+    taskId: string,
+    state: string,
+    agent?: string | null,
+    model?: string | null,
+    outputSummary?: string,
+  ): void {
+    void emitHook('turn.end', {
+      sessionId,
+      turnSummary: { taskId, state, agent, model: model ?? undefined, outputSummary },
+    });
+  }
+
   // F1 (OCE-001/OCE-002): finalize a streaming assistant message into a terminal
   // state and publish the matching message_complete frame. Best-effort + idempotent
   // (the helper's `WHERE status='streaming'` guard) — a failure here must never mask
@@ -318,6 +335,7 @@ export function createDispatcher(deps: Deps): {
 
     // Declared before try so the catch block can write it back on the task row.
     let chatId: string | null = null;
+    let sessionId: string | undefined;
 
     try {
       // Mark running
@@ -330,7 +348,6 @@ export function createDispatcher(deps: Deps): {
       // Session setup: reuse a pre-created session (e.g. Q&A arena contestants
       // whose persona is stamped on the session via agent_id) or create a fresh one.
       const model = task.model ?? config.DEFAULT_MODEL;
-      let sessionId: string;
       if (task.session_id) {
         sessionId = task.session_id;
       } else {
@@ -377,6 +394,7 @@ export function createDispatcher(deps: Deps): {
           SET state = 'cancelled', ended_at = clock_timestamp()
           WHERE id = ${taskId}
         `;
+        if (sessionId) emitTurnEnd(sessionId, taskId, 'cancelled', null, task.model);
         return;
       }
 
@@ -399,6 +417,7 @@ export function createDispatcher(deps: Deps): {
           WHERE id = ${taskId}
         `;
         log.info({ taskId, costTokens }, 'dispatcher: task completed (native)');
+        emitTurnEnd(sessionId, taskId, 'completed', null, task.model, summary);
       } else {
         const [msg] = await sql<{ content: string | null }[]>`
           SELECT content FROM messages WHERE id = ${assistantId}
@@ -410,6 +429,7 @@ export function createDispatcher(deps: Deps): {
           WHERE id = ${taskId}
         `;
         log.warn({ taskId, finalStatus }, 'dispatcher: task failed (native)');
+        emitTurnEnd(sessionId, taskId, 'failed', null, task.model, summary);
       }
     } catch (err) {
       const errMsg = err instanceof Error ? err.message : String(err);
@@ -419,6 +439,7 @@ export function createDispatcher(deps: Deps): {
         SET state = 'failed', ended_at = clock_timestamp(), output_summary = ${errMsg.slice(0, 500)}, chat_id = ${chatId}
         WHERE id = ${taskId}
       `.catch(() => {});
+      if (sessionId) emitTurnEnd(sessionId, taskId, 'failed', null, task.model, errMsg);
     }
   }
 
@@ -684,6 +705,7 @@ export function createDispatcher(deps: Deps): {
         await finalizeMessage(sessionId, chatId, assistantId, 'cancelled', task.model, assistantContent);
         await sql`UPDATE tasks SET state = 'cancelled', ended_at = clock_timestamp() WHERE id = ${taskId}`;
         emitAgentStatus(sessionId, chatId, agent, 'idle', stopping ? 'shutdown' : 'cancelled');
+        emitTurnEnd(sessionId, taskId, 'cancelled', agent, task.model);
         await cleanupWorktree(projectPath, taskId);
         clearTaskCommands(taskId);
         return;
@@ -738,6 +760,7 @@ export function createDispatcher(deps: Deps): {
       log.info({ taskId, agent, costTokens: extCostTokens }, 'dispatcher: task completed (external)');
       // #10: external-agent turn completed cleanly.
       emitAgentStatus(sessionId, chatId, agent, 'idle', 'turn_complete');
+      emitTurnEnd(sessionId, taskId, 'completed', agent, task.model, outputSummary);
       clearTaskCommands(taskId);
 
     } catch (err) {
@@ -762,6 +785,7 @@ export function createDispatcher(deps: Deps): {
       // preceded its assignment — guard so the status publish never masks the real
       // error.
       if (chatId) emitAgentStatus(sessionId, chatId, agent, status === 'cancelled' ? 'idle' : 'error', status === 'cancelled' ? 'cancelled' : 'failed');
+      if (sessionId) emitTurnEnd(sessionId, taskId, status, agent, task.model, errMsg);
 
       // Best-effort cleanup
       await cleanupWorktree(projectPath, taskId);
@@ -1030,6 +1054,7 @@ export function createDispatcher(deps: Deps): {
         await finalizeMessage(sessionId, chatId, assistantId, 'cancelled', task.model, assistantContent);
         await sql`UPDATE tasks SET state = 'cancelled', ended_at = clock_timestamp() WHERE id = ${taskId}`;
         emitAgentStatus(sessionId, chatId, agent, 'idle', stopping ? 'shutdown' : 'cancelled');
+        emitTurnEnd(sessionId, taskId, 'cancelled', agent, task.model);
         clearTaskCommands(taskId);
         return; // worktree persists (no cleanup); backend stays warm
       }
@@ -1090,6 +1115,7 @@ export function createDispatcher(deps: Deps): {
         result.ok ? 'idle' : 'error',
         result.ok ? 'turn_complete' : 'failed',
       );
+      emitTurnEnd(sessionId, taskId, finalState, agent, task.model, outputSummary);
       clearTaskCommands(taskId);
     } catch (err) {
       const errMsg = err instanceof Error ? err.message : String(err);
@@ -1104,6 +1130,7 @@ export function createDispatcher(deps: Deps): {
       await finalizeMessage(sessionId, chatId, assistantId, status, task.model);
       // #10: turn crashed.
       if (chatId) emitAgentStatus(sessionId, chatId, agent, status === 'cancelled' ? 'idle' : 'error', status === 'cancelled' ? 'cancelled' : 'crashed');
+      if (sessionId) emitTurnEnd(sessionId, taskId, status, agent, task.model, errMsg);
       clearTaskCommands(taskId);
       // No worktree cleanup (persistent); backend stays warm for the next turn.
     }
@@ -1308,6 +1335,7 @@ export function createDispatcher(deps: Deps): {
         await finalizeMessage(sessionId, chatId, assistantId, 'cancelled', task.model, assistantContent);
         await sql`UPDATE tasks SET state = 'cancelled', ended_at = clock_timestamp() WHERE id = ${taskId}`;
         emitAgentStatus(sessionId, chatId, agent, 'idle', stopping ? 'shutdown' : 'cancelled');
+        emitTurnEnd(sessionId, taskId, 'cancelled', agent, task.model);
         clearTaskCommands(taskId);
         return; // worktree persists (no cleanup); backend stays warm
       }
@@ -1367,6 +1395,7 @@ export function createDispatcher(deps: Deps): {
         result.ok ? 'idle' : 'error',
         result.ok ? 'turn_complete' : 'failed',
       );
+      emitTurnEnd(sessionId, taskId, finalState, agent, task.model, outputSummary);
       clearTaskCommands(taskId);
     } catch (err) {
       const errMsg = err instanceof Error ? err.message : String(err);
@@ -1381,6 +1410,7 @@ export function createDispatcher(deps: Deps): {
       await finalizeMessage(sessionId, chatId, assistantId, status, task.model);
       // #10: turn crashed.
       emitAgentStatus(sessionId, chatId, agent, status === 'cancelled' ? 'idle' : 'error', status === 'cancelled' ? 'cancelled' : 'crashed');
+      emitTurnEnd(sessionId, taskId, status, agent, task.model, errMsg);
       clearTaskCommands(taskId);
       // No worktree cleanup (persistent); backend stays warm for the next turn.
     }
@@ -1576,6 +1606,7 @@ export function createDispatcher(deps: Deps): {
         await finalizeMessage(sessionId, chatId, assistantId, 'cancelled', task.model, assistantContent);
         await sql`UPDATE tasks SET state = 'cancelled', ended_at = clock_timestamp() WHERE id = ${taskId}`;
         emitAgentStatus(sessionId, chatId, agent, 'idle', stopping ? 'shutdown' : 'cancelled');
+        emitTurnEnd(sessionId, taskId, 'cancelled', agent, task.model);
         clearTaskCommands(taskId);
         return; // worktree persists (no cleanup); backend stays warm
       }
@@ -1638,6 +1669,7 @@ export function createDispatcher(deps: Deps): {
         result.ok ? 'idle' : 'error',
         result.ok ? 'turn_complete' : 'failed',
       );
+      emitTurnEnd(sessionId, taskId, finalState, agent, task.model, outputSummary);
       clearTaskCommands(taskId);
     } catch (err) {
       const errMsg = err instanceof Error ? err.message : String(err);
@@ -1652,6 +1684,7 @@ export function createDispatcher(deps: Deps): {
       await finalizeMessage(sessionId, chatId, assistantId, status, task.model);
       // #10: turn crashed.
       emitAgentStatus(sessionId, chatId, agent, status === 'cancelled' ? 'idle' : 'error', status === 'cancelled' ? 'cancelled' : 'crashed');
+      emitTurnEnd(sessionId, taskId, status, agent, task.model, errMsg);
       clearTaskCommands(taskId);
       // No worktree cleanup (persistent); backend stays warm for the next turn.
     }

apps/coder/src/services/frame-emitter.ts

@@ -19,9 +19,10 @@
 import type { Broker } from '@boocode/server/broker';
 import type { WsFrame } from '@boocode/contracts/ws-frames';
 import type { AgentEvent } from './agent-backend.js';
-import { type AcpToolSnapshot, snapshotToWireToolCall } from './acp-tool-snapshot.js';
+import { type AcpToolSnapshot, snapshotToWireToolCall, mapToolLifecycleStatus } from './acp-tool-snapshot.js';
 import { mergeTaskCommands, getTaskCommands } from './agent-commands-cache.js';
 import type { DcpStreamStripper } from './dcp-strip.js';
+import { emitHook } from '../plugins/host.js';
 
 export interface FrameEmitterOpts {
   broker?: Broker;
@@ -91,8 +92,29 @@ export function makeFrameEmitter(opts: FrameEmitterOpts): FrameEmitter {
         }
         break;
       case 'tool_call':
+        toolSnapshots.set(e.toolCall.toolCallId, e.toolCall);
+        if (canStream()) {
+          broker!.publishFrame(sessionId!, {
+            type: 'tool_call',
+            message_id: assistantId!,
+            chat_id: chatId!,
+            tool_call: snapshotToWireToolCall(e.toolCall),
+          } as WsFrame);
+        }
+        break;
       case 'tool_update':
         toolSnapshots.set(e.toolCall.toolCallId, e.toolCall);
+        {
+          const lifecycle = mapToolLifecycleStatus(e.toolCall.status, e.toolCall.rawOutput);
+          if (lifecycle === 'completed' || lifecycle === 'failed') {
+            void emitHook('tool.execute.after', {
+              toolName: e.toolCall.title,
+              args: e.toolCall.rawInput,
+              result: e.toolCall.rawOutput,
+              duration: undefined,
+            });
+          }
+        }
         if (canStream()) {
           broker!.publishFrame(sessionId!, {
             type: 'tool_call',

apps/coder/src/services/hashline/constants.ts

@@ -0,0 +1,10 @@
+export const NIBBLE_STR = "ZPMQVRWSNKTXJBYH"
+
+export const HASHLINE_DICT = Array.from({ length: 256 }, (_, i) => {
+  const high = i >>> 4
+  const low = i & 0x0f
+  return `${NIBBLE_STR[high]}${NIBBLE_STR[low]}`
+})
+
+export const HASHLINE_REF_PATTERN = /^([0-9]+)#([ZPMQVRWSNKTXJBYH]{2})$/
+export const HASHLINE_OUTPUT_PATTERN = /^([0-9]+)#([ZPMQVRWSNKTXJBYH]{2})\|(.*)$/

apps/coder/src/services/hashline/hash-computation.ts

@@ -0,0 +1,31 @@
+import { HASHLINE_DICT } from "./constants.js"
+import { hashXxh32 } from "./xxhash32.js"
+
+const RE_SIGNIFICANT = /[\p{L}\p{N}]/u
+
+function computeNormalizedLineHash(lineNumber: number, normalizedContent: string): string {
+  const stripped = normalizedContent
+  const seed = RE_SIGNIFICANT.test(stripped) ? 0 : lineNumber
+  const hash = hashXxh32(stripped, seed)
+  const index = hash % 256
+  return HASHLINE_DICT[index]!
+}
+
+export function computeLineHash(lineNumber: number, content: string): string {
+  return computeNormalizedLineHash(lineNumber, content.replace(/\r/g, "").trimEnd())
+}
+
+export function computeLegacyLineHash(lineNumber: number, content: string): string {
+  return computeNormalizedLineHash(lineNumber, content.replace(/\r/g, "").replace(/\s+/g, ""))
+}
+
+export function formatHashLine(lineNumber: number, content: string): string {
+  const hash = computeLineHash(lineNumber, content)
+  return `${lineNumber}#${hash}|${content}`
+}
+
+export function formatHashLines(content: string): string {
+  if (!content) return ""
+  const lines = content.split("\n")
+  return lines.map((line, index) => formatHashLine(index + 1, line)).join("\n")
+}

apps/coder/src/services/hashline/index.ts

@@ -0,0 +1,11 @@
+/**
+ * Hashline editing core — content-hash anchors for edit_file stale-patch detection.
+ *
+ * Ported from oh-my-openagent/packages/hashline-core/.
+ * Bundles a runtime-aware xxHash32 (Bun fast-path, pure-JS fallback).
+ */
+export { computeLineHash, formatHashLines, formatHashLine, computeLegacyLineHash } from "./hash-computation.js"
+export { parseLineRef, validateLineRef, validateLineRefs, HashlineMismatchError, normalizeLineRef } from "./validation.js"
+export type { LineRef } from "./validation.js"
+export { NIBBLE_STR, HASHLINE_DICT, HASHLINE_REF_PATTERN, HASHLINE_OUTPUT_PATTERN } from "./constants.js"
+export type { ReplaceEdit, AppendEdit, PrependEdit, HashlineEdit } from "./types.js"

apps/coder/src/services/hashline/types.ts

@@ -0,0 +1,20 @@
+export interface ReplaceEdit {
+  op: "replace"
+  pos: string
+  end?: string
+  lines: string | string[]
+}
+
+export interface AppendEdit {
+  op: "append"
+  pos?: string
+  lines: string | string[]
+}
+
+export interface PrependEdit {
+  op: "prepend"
+  pos?: string
+  lines: string | string[]
+}
+
+export type HashlineEdit = ReplaceEdit | AppendEdit | PrependEdit

apps/coder/src/services/hashline/validation.ts

@@ -0,0 +1,192 @@
+import { computeLegacyLineHash, computeLineHash } from "./hash-computation.js"
+import { HASHLINE_REF_PATTERN } from "./constants.js"
+
+export interface LineRef {
+  line: number
+  hash: string
+}
+
+interface HashMismatch {
+  line: number
+  expected: string
+}
+
+const MISMATCH_CONTEXT = 2
+
+const LINE_REF_EXTRACT_PATTERN = /([0-9]+#[ZPMQVRWSNKTXJBYH]{2})/
+
+function isCompatibleLineHash(line: number, content: string, hash: string): boolean {
+  return computeLineHash(line, content) === hash || computeLegacyLineHash(line, content) === hash
+}
+
+export function normalizeLineRef(ref: string): string {
+  const originalTrimmed = ref.trim()
+  let trimmed = originalTrimmed
+  trimmed = trimmed.replace(/^(?:>>>|[+-])\s*/, "")
+  trimmed = trimmed.replace(/\s*#\s*/, "#")
+  trimmed = trimmed.replace(/\|.*$/, "")
+  trimmed = trimmed.trim()
+
+  if (HASHLINE_REF_PATTERN.test(trimmed)) {
+    return trimmed
+  }
+
+  const extracted = trimmed.match(LINE_REF_EXTRACT_PATTERN)
+  if (extracted) {
+    return extracted[1]!
+  }
+
+  return originalTrimmed
+}
+
+export function parseLineRef(ref: string): LineRef {
+  const normalized = normalizeLineRef(ref)
+  const match = normalized.match(HASHLINE_REF_PATTERN)
+  if (match) {
+    return {
+      line: Number.parseInt(match[1]!, 10),
+      hash: match[2]!,
+    }
+  }
+  const hashIdx = normalized.indexOf('#')
+  if (hashIdx > 0) {
+    const prefix = normalized.slice(0, hashIdx)
+    const suffix = normalized.slice(hashIdx + 1)
+    if (!/^\d+$/.test(prefix) && /^[ZPMQVRWSNKTXJBYH]{2}$/.test(suffix)) {
+      throw new Error(
+        `Invalid line reference: "${ref}". "${prefix}" is not a line number. ` +
+          `Use the actual line number from the read output.`
+      )
+    }
+  }
+  throw new Error(
+    `Invalid line reference format: "${ref}". Expected format: "{line_number}#{hash_id}"`
+  )
+}
+
+export function validateLineRef(lines: string[], ref: string): void {
+  const { line, hash } = parseLineRefWithHint(ref, lines)
+
+  if (line < 1 || line > lines.length) {
+    throw new Error(
+      `Line number ${line} out of bounds. File has ${lines.length} lines.`
+    )
+  }
+
+  const content = lines[line - 1]
+  if (content === undefined) {
+    throw new Error(
+      `Line number ${line} out of bounds. File has ${lines.length} lines.`
+    )
+  }
+  if (!isCompatibleLineHash(line, content, hash)) {
+    throw new HashlineMismatchError([{ line, expected: hash }], lines)
+  }
+}
+
+export class HashlineMismatchError extends Error {
+  readonly remaps: ReadonlyMap<string, string>
+
+  constructor(
+    private readonly mismatches: HashMismatch[],
+    private readonly fileLines: string[]
+  ) {
+    super(HashlineMismatchError.formatMessage(mismatches, fileLines))
+    this.name = "HashlineMismatchError"
+    const remaps = new Map<string, string>()
+    for (const mismatch of mismatches) {
+      const content = fileLines[mismatch.line - 1]
+      const actualLine = content ?? ""
+      const actual = computeLineHash(mismatch.line, actualLine)
+      remaps.set(`${mismatch.line}#${mismatch.expected}`, `${mismatch.line}#${actual}`)
+    }
+    this.remaps = remaps
+  }
+
+  static formatMessage(mismatches: HashMismatch[], fileLines: string[]): string {
+    const mismatchByLine = new Map<number, HashMismatch>()
+    for (const mismatch of mismatches) mismatchByLine.set(mismatch.line, mismatch)
+
+    const displayLines = new Set<number>()
+    for (const mismatch of mismatches) {
+      const low = Math.max(1, mismatch.line - MISMATCH_CONTEXT)
+      const high = Math.min(fileLines.length, mismatch.line + MISMATCH_CONTEXT)
+      for (let line = low; line <= high; line++) displayLines.add(line)
+    }
+
+    const sortedLines = [...displayLines].sort((a, b) => a - b)
+    const output: string[] = []
+    output.push(
+      `${mismatches.length} line${mismatches.length > 1 ? "s have" : " has"} changed since last read. ` +
+        "Use updated {line_number}#{hash_id} references below (>>> marks changed lines)."
+    )
+    output.push("")
+
+    let previousLine = -1
+    for (const line of sortedLines) {
+      if (previousLine !== -1 && line > previousLine + 1) {
+        output.push("    ...")
+      }
+      previousLine = line
+
+      const content = fileLines[line - 1] ?? ""
+      const hash = computeLineHash(line, content)
+      const prefix = `${line}#${hash}|${content}`
+      if (mismatchByLine.has(line)) {
+        output.push(`>>> ${prefix}`)
+      } else {
+        output.push(`    ${prefix}`)
+      }
+    }
+
+    return output.join("\n")
+  }
+}
+
+function suggestLineForHash(ref: string, lines: string[]): string | null {
+  const hashMatch = ref.trim().match(/#([ZPMQVRWSNKTXJBYH]{2})$/)
+  if (!hashMatch) return null
+  const hash = hashMatch[1]!
+  for (let i = 0; i < lines.length; i++) {
+    if (isCompatibleLineHash(i + 1, lines[i] ?? "", hash)) {
+      return `Did you mean "${i + 1}#${computeLineHash(i + 1, lines[i] ?? "")}"?`
+    }
+  }
+  return null
+}
+
+function parseLineRefWithHint(ref: string, lines: string[]): LineRef {
+  try {
+    return parseLineRef(ref)
+  } catch (parseError) {
+    const hint = suggestLineForHash(ref, lines)
+    if (hint && parseError instanceof Error) {
+      throw new Error(`${parseError.message} ${hint}`)
+    }
+    throw parseError
+  }
+}
+
+export function validateLineRefs(lines: string[], refs: string[]): void {
+  const mismatches: HashMismatch[] = []
+
+  for (const ref of refs) {
+    const { line, hash } = parseLineRefWithHint(ref, lines)
+
+    if (line < 1 || line > lines.length) {
+      throw new Error(`Line number ${line} out of bounds (file has ${lines.length} lines)`)
+    }
+
+    const content = lines[line - 1]
+    if (content === undefined) {
+      throw new Error(`Line number ${line} out of bounds (file has ${lines.length} lines)`)
+    }
+    if (!isCompatibleLineHash(line, content, hash)) {
+      mismatches.push({ line, expected: hash })
+    }
+  }
+
+  if (mismatches.length > 0) {
+    throw new HashlineMismatchError(mismatches, lines)
+  }
+}

apps/coder/src/services/hashline/xxhash32.ts

@@ -0,0 +1,90 @@
+type BunHashRuntime = { hash: { xxHash32(data: string | Uint8Array, seed: number): number } }
+
+const runtime = globalThis as typeof globalThis & { Bun?: BunHashRuntime }
+const encoder = new TextEncoder()
+
+const PRIME32_1 = 0x9e3779b1
+const PRIME32_2 = 0x85ebca77
+const PRIME32_3 = 0xc2b2ae3d
+const PRIME32_4 = 0x27d4eb2f
+const PRIME32_5 = 0x165667b1
+
+function rotateLeft32(value: number, bits: number): number {
+  return ((value << bits) | (value >>> (32 - bits))) >>> 0
+}
+
+function readUint32LittleEndian(input: Uint8Array, offset: number): number {
+  return (
+    ((input[offset] ?? 0) |
+      ((input[offset + 1] ?? 0) << 8) |
+      ((input[offset + 2] ?? 0) << 16) |
+      ((input[offset + 3] ?? 0) << 24)) >>>
+    0
+  )
+}
+
+function round32(accumulator: number, value: number): number {
+  const added = (accumulator + Math.imul(value, PRIME32_2)) >>> 0
+  return Math.imul(rotateLeft32(added, 13), PRIME32_1) >>> 0
+}
+
+function xxHash32Js(input: Uint8Array, seed: number): number {
+  let offset = 0
+  const length = input.length
+  let hash: number
+
+  if (length >= 16) {
+    const limit = length - 16
+    let value1 = (seed + PRIME32_1 + PRIME32_2) >>> 0
+    let value2 = (seed + PRIME32_2) >>> 0
+    let value3 = seed >>> 0
+    let value4 = (seed - PRIME32_1) >>> 0
+
+    while (offset <= limit) {
+      value1 = round32(value1, readUint32LittleEndian(input, offset))
+      offset += 4
+      value2 = round32(value2, readUint32LittleEndian(input, offset))
+      offset += 4
+      value3 = round32(value3, readUint32LittleEndian(input, offset))
+      offset += 4
+      value4 = round32(value4, readUint32LittleEndian(input, offset))
+      offset += 4
+    }
+
+    hash = (rotateLeft32(value1, 1) + rotateLeft32(value2, 7)) >>> 0
+    hash = (hash + rotateLeft32(value3, 12)) >>> 0
+    hash = (hash + rotateLeft32(value4, 18)) >>> 0
+  } else {
+    hash = (seed + PRIME32_5) >>> 0
+  }
+
+  hash = (hash + length) >>> 0
+
+  while (offset + 4 <= length) {
+    hash = (hash + Math.imul(readUint32LittleEndian(input, offset), PRIME32_3)) >>> 0
+    hash = Math.imul(rotateLeft32(hash, 17), PRIME32_4) >>> 0
+    offset += 4
+  }
+
+  while (offset < length) {
+    hash = (hash + Math.imul(input[offset] ?? 0, PRIME32_5)) >>> 0
+    hash = Math.imul(rotateLeft32(hash, 11), PRIME32_1) >>> 0
+    offset += 1
+  }
+
+  hash = (hash ^ (hash >>> 15)) >>> 0
+  hash = Math.imul(hash, PRIME32_2) >>> 0
+  hash = (hash ^ (hash >>> 13)) >>> 0
+  hash = Math.imul(hash, PRIME32_3) >>> 0
+
+  return (hash ^ (hash >>> 16)) >>> 0
+}
+
+export function hashXxh32(input: string, seed: number): number {
+  const bun = runtime.Bun
+  if (bun !== undefined) {
+    return bun.hash.xxHash32(input, seed)
+  }
+
+  return xxHash32Js(encoder.encode(input), seed >>> 0)
+}

apps/coder/src/services/model-resolution/connected-providers-cache.ts

@@ -0,0 +1,34 @@
+import type { ModelMetadata } from "./provider-cache.js"
+
+export interface ProviderModelsCache {
+  readonly models: Record<string, readonly string[] | readonly ModelMetadata[]>
+  readonly connected: readonly string[]
+  readonly updatedAt: string
+}
+
+export interface ConnectedProvidersAdapter {
+  readConnectedProvidersCache(): string[] | null
+  findProviderModelMetadata(providerID: string, modelID: string): ModelMetadata | undefined
+  readProviderModelsCache(): ProviderModelsCache | null
+}
+
+export function readConnectedProvidersCache(): string[] | null {
+  return null
+}
+
+export function findProviderModelMetadata(
+  _providerID: string,
+  _modelID: string,
+): ModelMetadata | undefined {
+  return undefined
+}
+
+export function readProviderModelsCache(): ProviderModelsCache | null {
+  return null
+}
+
+export const connectedProvidersAdapter: ConnectedProvidersAdapter = {
+  readConnectedProvidersCache,
+  findProviderModelMetadata,
+  readProviderModelsCache,
+}

apps/coder/src/services/model-resolution/fallback-chain-from-models.ts

@@ -0,0 +1,128 @@
+import type { FallbackEntry } from "./model-requirement-types.js"
+import type { FallbackModelObject } from "./fallback-model-object.js"
+import { normalizeFallbackModels } from "./model-resolver.js"
+import { KNOWN_VARIANTS } from "./known-variants.js"
+
+function parseVariantFromModel(rawModel: string): { modelID: string; variant?: string } {
+  if (typeof rawModel !== "string") {
+    return { modelID: "" }
+  }
+  const trimmedModel = rawModel.trim()
+  if (!trimmedModel) {
+    return { modelID: "" }
+  }
+
+  const parenthesizedVariant = trimmedModel.match(/^(.*)\(([^()]+)\)\s*$/)
+  if (parenthesizedVariant) {
+    const modelID = parenthesizedVariant[1]?.trim() ?? ""
+    const variant = parenthesizedVariant[2]?.trim()
+    return variant ? { modelID, variant } : { modelID }
+  }
+
+  const spaceVariant = trimmedModel.match(/^(.*\S)\s+([a-z][a-z0-9_-]*)$/i)
+  if (spaceVariant) {
+    const modelID = spaceVariant[1]?.trim() ?? ""
+    const variant = spaceVariant[2]?.trim().toLowerCase()
+    if (variant && KNOWN_VARIANTS.has(variant)) {
+      return { modelID, variant }
+    }
+  }
+
+  return { modelID: trimmedModel }
+}
+
+export function parseFallbackModelEntry(
+  model: string,
+  contextProviderID: string | undefined,
+  defaultProviderID = "opencode",
+): FallbackEntry | undefined {
+  if (typeof model !== "string") return undefined
+  const trimmed = model.trim()
+  if (!trimmed) return undefined
+
+  const parts = trimmed.split("/")
+  const providerID =
+    parts.length >= 2 ? (parts[0]?.trim() ?? "") : (contextProviderID?.trim() || defaultProviderID)
+  const rawModelID = parts.length >= 2 ? parts.slice(1).join("/").trim() : trimmed
+  if (!providerID || !rawModelID) return undefined
+
+  const parsed = parseVariantFromModel(rawModelID)
+  if (!parsed.modelID) return undefined
+
+  return {
+    providers: [providerID],
+    model: parsed.modelID,
+    variant: parsed.variant,
+  }
+}
+
+export function parseFallbackModelObjectEntry(
+  obj: FallbackModelObject,
+  contextProviderID: string | undefined,
+  defaultProviderID = "opencode",
+): FallbackEntry | undefined {
+  const base = parseFallbackModelEntry(obj.model, contextProviderID, defaultProviderID)
+  if (!base) return undefined
+
+  return {
+    ...base,
+    variant: obj.variant ?? base.variant,
+    reasoningEffort: obj.reasoningEffort,
+    temperature: obj.temperature,
+    top_p: obj.top_p,
+    maxTokens: obj.maxTokens,
+    thinking: obj.thinking,
+  }
+}
+
+/**
+ * Find the most specific FallbackEntry whose `provider/model` is a prefix of
+ * the resolved `provider/modelID`.  Longest match wins so that e.g.
+ * `openai/gpt-5.4-preview` picks the entry for `openai/gpt-5.4-preview` over
+ * the shorter `openai/gpt-5.4`.
+ */
+export function findMostSpecificFallbackEntry(
+  providerID: string,
+  modelID: string,
+  chain: FallbackEntry[],
+): FallbackEntry | undefined {
+  const resolved = `${providerID}/${modelID}`.toLowerCase()
+
+  // Collect entries whose provider/model is a prefix of the resolved model,
+  // together with the length of the matching prefix (longest match wins).
+  const matches: { entry: FallbackEntry; matchLen: number }[] = []
+  for (const entry of chain) {
+    for (const p of entry.providers) {
+      const candidate = `${p}/${entry.model}`.toLowerCase()
+      if (resolved.startsWith(candidate)) {
+        matches.push({ entry, matchLen: candidate.length })
+        break // one match per entry is enough
+      }
+    }
+  }
+
+  if (matches.length === 0) return undefined
+  matches.sort((a, b) => b.matchLen - a.matchLen)
+  return matches[0]!.entry
+}
+
+export function buildFallbackChainFromModels(
+  fallbackModels: string | (string | FallbackModelObject)[] | undefined,
+  contextProviderID: string | undefined,
+  defaultProviderID = "opencode",
+): FallbackEntry[] | undefined {
+  const normalized = normalizeFallbackModels(fallbackModels)
+  if (!normalized || normalized.length === 0) return undefined
+
+  const parsed = normalized
+    .map((entry) => {
+      if (typeof entry === "string") {
+        return parseFallbackModelEntry(entry, contextProviderID, defaultProviderID)
+      }
+      return parseFallbackModelObjectEntry(entry, contextProviderID, defaultProviderID)
+    })
+    .filter((entry): entry is FallbackEntry => entry !== undefined)
+
+  if (parsed.length === 0) return undefined
+  return parsed
+}

apps/coder/src/services/model-resolution/fallback-model-object.ts

@@ -0,0 +1,9 @@
+export type FallbackModelObject = {
+  readonly model: string
+  readonly variant?: string
+  readonly reasoningEffort?: "none" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max"
+  readonly temperature?: number
+  readonly top_p?: number
+  readonly maxTokens?: number
+  readonly thinking?: { readonly type: "enabled" | "disabled"; readonly budgetTokens?: number }
+}

apps/coder/src/services/model-resolution/index.ts

@@ -0,0 +1,80 @@
+export type {
+  FallbackEntry,
+  ModelRequirement,
+} from "./model-requirement-types.js"
+export type {
+  FallbackModelObject,
+} from "./fallback-model-object.js"
+export type {
+  DelegatedModelConfig,
+  ModelResolutionRequest,
+  ModelResolutionProvenance,
+  ModelResolutionResult,
+} from "./model-resolution-types.js"
+export type {
+  ModelResolutionInput,
+  ModelSource,
+  ExtendedModelResolutionInput,
+} from "./model-resolver.js"
+export {
+  resolveModel,
+  resolveModelWithFallback,
+  normalizeFallbackModels,
+  flattenToFallbackModelStrings,
+} from "./model-resolver.js"
+export {
+  normalizeModel,
+  normalizeModelID,
+} from "./model-normalization.js"
+export {
+  fuzzyMatchModel,
+  isModelAvailable,
+} from "./model-availability.js"
+export {
+  transformModelForProvider,
+  transformModelForProviderDisplay,
+} from "./provider-model-id-transform.js"
+export {
+  buildFallbackChainFromModels,
+  parseFallbackModelEntry,
+  parseFallbackModelObjectEntry,
+  findMostSpecificFallbackEntry,
+} from "./fallback-chain-from-models.js"
+export {
+  KNOWN_VARIANTS,
+} from "./known-variants.js"
+export {
+  _setModelResolutionLogImplementationForTesting,
+  resolveModelPipeline,
+} from "./model-resolution-pipeline.js"
+export type {
+  ModelResolutionRequest as PipelineModelResolutionRequest,
+  ModelResolutionProvenance as PipelineModelResolutionProvenance,
+  ModelResolutionResult as PipelineModelResolutionResult,
+  ModelResolutionDeps,
+} from "./model-resolution-pipeline.js"
+export {
+  isRetryableModelError,
+  shouldRetryError,
+  getNextFallback,
+  hasMoreFallbacks,
+  selectFallbackProvider,
+  selectFallbackProviderWithCache,
+} from "./model-error-classifier.js"
+export type {
+  ErrorInfo,
+} from "./model-error-classifier.js"
+export type {
+  ProviderCache,
+  ModelMetadata,
+} from "./provider-cache.js"
+export type {
+  ProviderModelsCache,
+  ConnectedProvidersAdapter,
+} from "./connected-providers-cache.js"
+export {
+  readConnectedProvidersCache,
+  findProviderModelMetadata,
+  readProviderModelsCache,
+  connectedProvidersAdapter,
+} from "./connected-providers-cache.js"

apps/coder/src/services/model-resolution/known-variants.ts

@@ -0,0 +1,16 @@
+/**
+ * Canonical set of recognised variant / effort tokens.
+ * Used by parseFallbackModelEntry (space-suffix detection) and
+ * flattenToFallbackModelStrings (inline-variant stripping).
+ */
+export const KNOWN_VARIANTS = new Set([
+  "low",
+  "medium",
+  "high",
+  "xhigh",
+  "max",
+  "minimal",
+  "none",
+  "auto",
+  "thinking",
+])

apps/coder/src/services/model-resolution/model-availability.ts

@@ -0,0 +1,64 @@
+function normalizeModelName(name: string): string {
+  return name
+    .toLowerCase()
+    .replace(/claude-(opus|sonnet|haiku)-(\d+)[.-](\d+)/g, "claude-$1-$2.$3")
+}
+
+export function fuzzyMatchModel(
+  target: string,
+  available: Set<string>,
+  providers?: string[],
+): string | null {
+  if (available.size === 0) {
+    return null
+  }
+
+  const targetNormalized = normalizeModelName(target)
+
+  let candidates = Array.from(available)
+  if (providers && providers.length > 0) {
+    const providerSet = new Set(providers)
+    candidates = candidates.filter((model) => {
+      const [provider] = model.split("/")
+      return providerSet.has(provider!)
+    })
+  }
+
+  if (candidates.length === 0) {
+    return null
+  }
+
+  const matches = candidates.filter((model) =>
+    normalizeModelName(model).includes(targetNormalized),
+  )
+
+  if (matches.length === 0) {
+    return null
+  }
+
+  const exactMatch = matches.find((model) => normalizeModelName(model) === targetNormalized)
+  if (exactMatch) {
+    return exactMatch
+  }
+
+  const exactModelIdMatches = matches.filter((model) => {
+    const modelId = model.split("/").slice(1).join("/")
+    return normalizeModelName(modelId) === targetNormalized
+  })
+  if (exactModelIdMatches.length > 0) {
+    return exactModelIdMatches.reduce((shortest, current) =>
+      current.length < shortest.length ? current : shortest,
+    )
+  }
+
+  return matches.reduce((shortest, current) =>
+    current.length < shortest.length ? current : shortest,
+  )
+}
+
+export function isModelAvailable(
+  targetModel: string,
+  availableModels: Set<string>,
+): boolean {
+  return fuzzyMatchModel(targetModel, availableModels) !== null
+}

apps/coder/src/services/model-resolution/model-error-classifier.ts

@@ -0,0 +1,261 @@
+import type { FallbackEntry } from "./model-requirement-types.js"
+import type { ProviderCache } from "./provider-cache.js"
+import * as connectedProvidersCache from "./connected-providers-cache.js"
+
+/**
+ * Error names that indicate a retryable model error.
+ * These errors halt execution and should trigger fallback retry.
+ */
+const RETRYABLE_ERROR_NAMES = new Set([
+  "providermodelnotfounderror",
+  "ratelimiterror",
+  "modelunavailableerror",
+  "providerconnectionerror",
+  "authenticationerror",
+])
+
+const STOP_ERROR_NAMES = new Set([
+  "quotaexceedederror",
+  "insufficientcreditserror",
+  "freeusagelimiterror",
+])
+
+/**
+ * Error names that should NOT trigger retry.
+ * These errors are typically user-induced or fixable without switching models.
+ */
+const NON_RETRYABLE_ERROR_NAMES = new Set([
+  "messageabortederror",
+  "permissiondeniederror",
+  "contextlengtherror",
+  "timeouterror",
+  "validationerror",
+  "syntaxerror",
+  "usererror",
+])
+
+/**
+ * Message patterns that indicate a retryable error even without a known error name.
+ */
+const RETRYABLE_MESSAGE_PATTERNS = [
+  "rate_limit",
+  "rate limit",
+  "usage_limit_reached",
+  "usage limit has been reached",
+  "quota",
+  "all credentials for model",
+  "cooling down",
+  "exhausted your capacity",
+  "not found",
+  "unavailable",
+  "insufficient",
+  "too many requests",
+  "over limit",
+  "overloaded",
+  "bad gateway",
+  "bad request",
+  "unknown provider",
+  "provider not found",
+  "model_not_supported",
+  "model not supported",
+  "model is not supported",
+  "connection error",
+  "network error",
+  "timeout",
+  "service unavailable",
+  "internal_server_error",
+  "free usage",
+  "usage exceeded",
+  "credit",
+  "balance",
+  "temporarily unavailable",
+  "try again",
+  "请稍后重试",
+  "503",
+  "502",
+  "504",
+  "429",
+  "529",
+  "selected provider is forbidden",
+  "provider is forbidden",
+  // Chinese retryable patterns (Zhipu, etc.)
+  "频率限制",           // "rate limit"
+  "请求过于频繁",       // "too many requests"
+  "暂时不可用",         // "temporarily unavailable"
+  "服务不可用",         // "service unavailable"
+  "server_error",
+  "an error occurred while processing",
+]
+
+/**
+ * Message patterns that indicate a non-retryable STOP error (quota/billing exhaustion).
+ * These take precedence over RETRYABLE_MESSAGE_PATTERNS.
+ */
+const STOP_MESSAGE_PATTERNS = [
+  "quota will reset after",
+  "quota exceeded",
+  "free usage limit",
+  "billing limit",
+  "billing hard limit",
+  "monthly limit",
+  "plan limit",
+  "subscription quota",
+  "subscription limit",
+  "payment required",
+  "out of credits",
+  "credits exhausted",
+  "insufficient credits",
+  "insufficient balance",
+  "credit balance",
+  "usage limit for this month",
+  "exhausted your capacity",
+  // GLM/Z.ai business error codes that indicate permanent quota/billing exhaustion
+  "daily call limit",
+  "daily limit",
+  "usage limit reached for",
+  "in arrears",
+  "fair use policy",
+  "recharge and try",
+  "使用上限",
+  "额度不足",
+  "余额不足",
+  "已耗尽",
+]
+
+const AUTO_RETRY_GATE_PATTERNS = [
+  "rate limit",
+  "cooling down",
+  "credentials for model",
+]
+
+function hasProviderAutoRetrySignal(message: string): boolean {
+  if (!message.includes("retrying in")) {
+    return false
+  }
+  return AUTO_RETRY_GATE_PATTERNS.some((pattern) => message.includes(pattern))
+}
+
+export interface ErrorInfo {
+  name?: string
+  message?: string
+  /** HTTP status code from the provider response (e.g., 429 for rate limit) */
+  statusCode?: number
+}
+
+/**
+ * Determines if an error is a retryable model error.
+ * Returns true if it's a known retryable type OR matches retryable message patterns.
+ */
+export function isRetryableModelError(error: ErrorInfo): boolean {
+  // If we have an error name, check against known lists
+  if (error.name) {
+    const errorNameLower = error.name.toLowerCase()
+    // Explicit non-retryable takes precedence
+    if (NON_RETRYABLE_ERROR_NAMES.has(errorNameLower)) {
+      return false
+    }
+    if (STOP_ERROR_NAMES.has(errorNameLower)) {
+      return false
+    }
+    // Check if it's a known retryable error
+    if (RETRYABLE_ERROR_NAMES.has(errorNameLower)) {
+      return true
+    }
+  }
+
+  // Check message patterns for unknown errors
+  const msg = error.message?.toLowerCase() ?? ""
+
+  // STOP patterns take precedence over retryable patterns
+  if (STOP_MESSAGE_PATTERNS.some((pattern) => msg.includes(pattern))) {
+    return false
+  }
+
+  if (hasProviderAutoRetrySignal(msg)) {
+    return true
+  }
+
+  // HTTP status code check: catches rate-limit errors regardless of message format/language.
+  // Uses the same codes as runtime-fallback config (400 excluded as it is a permanent client error).
+  if (
+    error.statusCode != null &&
+    (error.statusCode === 429 || error.statusCode === 503 || error.statusCode === 529)
+  ) {
+    return true
+  }
+
+  return RETRYABLE_MESSAGE_PATTERNS.some((pattern) => msg.includes(pattern))
+}
+
+/**
+ * Determines if an error should trigger a fallback retry.
+ * Returns true for errors that halt execution.
+ */
+export function shouldRetryError(error: ErrorInfo): boolean {
+  return isRetryableModelError(error)
+}
+
+/**
+ * Gets the next fallback model from the chain based on attempt count.
+ * Returns undefined if all fallbacks have been exhausted.
+ */
+export function getNextFallback(
+  fallbackChain: FallbackEntry[],
+  attemptCount: number,
+): FallbackEntry | undefined {
+  return fallbackChain[attemptCount]
+}
+
+/**
+ * Checks if there are more fallbacks available after the current attempt.
+ */
+export function hasMoreFallbacks(
+  fallbackChain: FallbackEntry[],
+  attemptCount: number,
+): boolean {
+  return attemptCount < fallbackChain.length
+}
+
+/**
+ * Selects the best provider for a fallback entry.
+ * Priority:
+ * 1) First connected provider in the entry's provider preference order
+ * 2) Preferred provider when connected (and entry providers are unavailable)
+ * 3) First provider listed in the fallback entry
+ */
+export function selectFallbackProvider(
+  providers: string[],
+  preferredProviderID?: string,
+): string {
+  return selectFallbackProviderWithCache(
+    providers,
+    connectedProvidersCache,
+    preferredProviderID,
+  )
+}
+
+export function selectFallbackProviderWithCache(
+  providers: string[],
+  providerCache: ProviderCache,
+  preferredProviderID?: string,
+): string {
+  const connectedProviders = providerCache.readConnectedProvidersCache()
+  if (connectedProviders) {
+    const connectedSet = new Set(connectedProviders.map(p => p.toLowerCase()))
+
+    for (const provider of providers) {
+      if (connectedSet.has(provider.toLowerCase())) {
+        return provider
+      }
+    }
+
+    if (
+      preferredProviderID &&
+      connectedSet.has(preferredProviderID.toLowerCase())
+    ) {
+      return preferredProviderID
+    }
+  }
+
+  return providers[0] ?? preferredProviderID ?? "opencode"
+}

apps/coder/src/services/model-resolution/model-normalization.ts

@@ -0,0 +1,8 @@
+export function normalizeModel(model?: string): string | undefined {
+  const trimmed = model?.trim()
+  return trimmed || undefined
+}
+
+export function normalizeModelID(modelID: string): string {
+  return modelID.replace(/\.(\d+)/g, "-$1")
+}

apps/coder/src/services/model-resolution/model-requirement-types.ts

@@ -0,0 +1,18 @@
+export type FallbackEntry = {
+  providers: string[];
+  model: string;
+  variant?: string; // Entry-specific variant (e.g., GPT->high, Opus->max)
+  reasoningEffort?: string;
+  temperature?: number;
+  top_p?: number;
+  maxTokens?: number;
+  thinking?: { type: "enabled" | "disabled"; budgetTokens?: number };
+};
+
+export type ModelRequirement = {
+  fallbackChain: FallbackEntry[];
+  variant?: string; // Default variant (used when entry doesn't specify one)
+  requiresModel?: string; // If set, only activates when this model is available (fuzzy match)
+  requiresAnyModel?: boolean; // If true, requires at least ONE model in fallbackChain to be available (or empty availability treated as unavailable)
+  requiresProvider?: string[]; // If set, only activates when any of these providers is connected
+};

apps/coder/src/services/model-resolution/model-resolution-pipeline.ts

@@ -0,0 +1,256 @@
+import { fuzzyMatchModel } from "./model-availability.js"
+import type { FallbackEntry } from "./model-requirement-types.js"
+import { transformModelForProvider } from "./provider-model-id-transform.js"
+import { normalizeModel } from "./model-normalization.js"
+import type { ProviderCache } from "./provider-cache.js"
+
+type LogImplementation = (message: string, data?: unknown) => void
+
+let logImplementationForTesting: LogImplementation | undefined
+
+function log(message: string, data?: unknown): void {
+  const logImpl = logImplementationForTesting
+  if (!logImpl) {
+    return
+  }
+  if (arguments.length === 1) {
+    logImpl(message)
+    return
+  }
+  logImpl(message, data)
+}
+
+export function _setModelResolutionLogImplementationForTesting(
+  logImplementation: LogImplementation | undefined,
+): void {
+  logImplementationForTesting = logImplementation
+}
+
+export type ModelResolutionRequest = {
+  intent?: {
+    uiSelectedModel?: string
+    userModel?: string
+    userFallbackModels?: string[]
+    categoryDefaultModel?: string
+  }
+  constraints: {
+    availableModels: Set<string>
+    connectedProviders?: string[] | null
+  }
+  policy?: {
+    fallbackChain?: FallbackEntry[]
+    systemDefaultModel?: string
+  }
+}
+
+export type ModelResolutionProvenance =
+  | "override"
+  | "category-default"
+  | "provider-fallback"
+  | "system-default"
+
+export type ModelResolutionResult = {
+  model: string
+  provenance: ModelResolutionProvenance
+  variant?: string
+  attempted?: string[]
+  reason?: string
+}
+
+export type ModelResolutionDeps = {
+  fuzzyMatchModel: (
+    target: string,
+    available: Set<string>,
+    providers?: string[],
+  ) => string | null
+  transformModelForProvider: (provider: string, model: string) => string
+}
+
+const DEFAULT_MODEL_RESOLUTION_DEPS: ModelResolutionDeps = {
+  fuzzyMatchModel,
+  transformModelForProvider,
+}
+
+
+export function resolveModelPipeline(
+  request: ModelResolutionRequest,
+  providerCache: ProviderCache = {
+    readConnectedProvidersCache: () => null,
+    findProviderModelMetadata: () => undefined,
+  },
+  deps: ModelResolutionDeps = DEFAULT_MODEL_RESOLUTION_DEPS,
+): ModelResolutionResult | undefined {
+  const attempted: string[] = []
+  const { intent, constraints, policy } = request
+  const availableModels = constraints.availableModels
+  const fallbackChain = policy?.fallbackChain
+  const systemDefaultModel = policy?.systemDefaultModel
+
+  const normalizedUiModel = normalizeModel(intent?.uiSelectedModel)
+  if (normalizedUiModel) {
+    log("Model resolved via UI selection", { model: normalizedUiModel })
+    return { model: normalizedUiModel, provenance: "override" }
+  }
+
+  const normalizedUserModel = normalizeModel(intent?.userModel)
+  if (normalizedUserModel) {
+    log("Model resolved via config override", { model: normalizedUserModel })
+    return { model: normalizedUserModel, provenance: "override" }
+  }
+
+  const normalizedCategoryDefault = normalizeModel(intent?.categoryDefaultModel)
+  if (normalizedCategoryDefault) {
+    attempted.push(normalizedCategoryDefault)
+    if (availableModels.size > 0) {
+      const parts = normalizedCategoryDefault.split("/")
+      const providerHint = parts.length >= 2 ? [parts[0]!] : undefined
+      const match = deps.fuzzyMatchModel(normalizedCategoryDefault, availableModels, providerHint)
+      if (match) {
+        log("Model resolved via category default (fuzzy matched)", {
+          original: normalizedCategoryDefault,
+          matched: match,
+        })
+        return { model: match, provenance: "category-default", attempted }
+      }
+    } else {
+      const connectedProviders = constraints.connectedProviders ?? providerCache.readConnectedProvidersCache()
+      if (connectedProviders === null) {
+        log("Model resolved via category default (no cache, first run)", {
+          model: normalizedCategoryDefault,
+        })
+        return { model: normalizedCategoryDefault, provenance: "category-default", attempted }
+      }
+      const parts = normalizedCategoryDefault.split("/")
+      if (parts.length >= 2) {
+        const provider = parts[0]!
+        if (connectedProviders.includes(provider)) {
+          const modelName = parts.slice(1).join("/")
+          const transformedModel = `${provider}/${deps.transformModelForProvider(provider, modelName)}`
+          log("Model resolved via category default (connected provider)", {
+            model: transformedModel,
+            original: normalizedCategoryDefault,
+          })
+          return { model: transformedModel, provenance: "category-default", attempted }
+        }
+      }
+    }
+    log("Category default model not available, falling through to fallback chain", {
+      model: normalizedCategoryDefault,
+    })
+  }
+
+  //#when - user configured fallback_models, try them before hardcoded fallback chain
+  const userFallbackModels = intent?.userFallbackModels
+  if (userFallbackModels && userFallbackModels.length > 0) {
+    if (availableModels.size === 0) {
+      const connectedProviders = constraints.connectedProviders ?? providerCache.readConnectedProvidersCache()
+      const connectedSet = connectedProviders ? new Set(connectedProviders) : null
+
+      if (connectedSet !== null) {
+        for (const model of userFallbackModels) {
+          attempted.push(model)
+          const parts = model.split("/")
+          if (parts.length >= 2) {
+            const provider = parts[0]!
+            if (connectedSet.has(provider)) {
+              const modelName = parts.slice(1).join("/")
+              const transformedModel = `${provider}/${deps.transformModelForProvider(provider, modelName)}`
+              log("Model resolved via user fallback_models (connected provider)", { model: transformedModel, original: model })
+              return { model: transformedModel, provenance: "provider-fallback", attempted }
+            }
+          }
+        }
+        log("No connected provider found in user fallback_models, falling through to hardcoded chain")
+      }
+    } else {
+      for (const model of userFallbackModels) {
+        attempted.push(model)
+        const parts = model.split("/")
+        const providerHint = parts.length >= 2 ? [parts[0]!] : undefined
+        const match = deps.fuzzyMatchModel(model, availableModels, providerHint)
+        if (match) {
+          log("Model resolved via user fallback_models (availability confirmed)", { model, match })
+          return { model: match, provenance: "provider-fallback", attempted }
+        }
+      }
+      log("No available model found in user fallback_models, falling through to hardcoded chain")
+    }
+  }
+
+  if (fallbackChain && fallbackChain.length > 0) {
+    if (availableModels.size === 0) {
+      const connectedProviders = constraints.connectedProviders ?? providerCache.readConnectedProvidersCache()
+      const connectedSet = connectedProviders ? new Set(connectedProviders) : null
+
+      if (connectedSet === null) {
+        log("Model fallback chain skipped (no connected providers cache) - falling through to system default")
+      } else {
+        for (const entry of fallbackChain) {
+          for (const provider of entry.providers) {
+            if (connectedSet.has(provider)) {
+              const transformedModelId = deps.transformModelForProvider(provider, entry.model)
+              const model = `${provider}/${transformedModelId}`
+              log("Model resolved via fallback chain (connected provider)", {
+                provider,
+                model: transformedModelId,
+                variant: entry.variant,
+              })
+              return {
+                model,
+                provenance: "provider-fallback",
+                variant: entry.variant,
+                attempted,
+              }
+            }
+          }
+        }
+        log("No connected provider found in fallback chain, falling through to system default")
+      }
+    } else {
+      for (const entry of fallbackChain) {
+        for (const provider of entry.providers) {
+          const fullModel = `${provider}/${entry.model}`
+          const match = deps.fuzzyMatchModel(fullModel, availableModels, [provider])
+          if (match) {
+            log("Model resolved via fallback chain (availability confirmed)", {
+              provider,
+              model: entry.model,
+              match,
+              variant: entry.variant,
+            })
+            return {
+              model: match,
+              provenance: "provider-fallback",
+              variant: entry.variant,
+              attempted,
+            }
+          }
+        }
+
+        const crossProviderMatch = deps.fuzzyMatchModel(entry.model, availableModels)
+        if (crossProviderMatch) {
+          log("Model resolved via fallback chain (cross-provider fuzzy match)", {
+            model: entry.model,
+            match: crossProviderMatch,
+            variant: entry.variant,
+          })
+          return {
+            model: crossProviderMatch,
+            provenance: "provider-fallback",
+            variant: entry.variant,
+            attempted,
+          }
+        }
+      }
+      log("No available model found in fallback chain, falling through to system default")
+    }
+  }
+
+  if (systemDefaultModel === undefined) {
+    log("No model resolved - systemDefaultModel not configured")
+    return undefined
+  }
+
+  log("Model resolved via system default", { model: systemDefaultModel })
+  return { model: systemDefaultModel, provenance: "system-default", attempted }
+}

apps/coder/src/services/model-resolution/model-resolution-types.ts

@@ -0,0 +1,41 @@
+import type { FallbackEntry } from "./model-requirement-types.js"
+
+export interface DelegatedModelConfig {
+  providerID: string
+  modelID: string
+  variant?: string
+  reasoningEffort?: string
+  temperature?: number
+  top_p?: number
+  maxTokens?: number
+  thinking?: { type: "enabled" | "disabled"; budgetTokens?: number }
+}
+
+export type ModelResolutionRequest = {
+  intent?: {
+    uiSelectedModel?: string
+    userModel?: string
+    categoryDefaultModel?: string
+  }
+  constraints: {
+    availableModels: Set<string>
+  }
+  policy?: {
+    fallbackChain?: FallbackEntry[]
+    systemDefaultModel?: string
+  }
+}
+
+export type ModelResolutionProvenance =
+  | "override"
+  | "category-default"
+  | "provider-fallback"
+  | "system-default"
+
+export type ModelResolutionResult = {
+  model: string
+  provenance: ModelResolutionProvenance
+  variant?: string
+  attempted?: string[]
+  reason?: string
+}

apps/coder/src/services/model-resolution/model-resolver.ts

@@ -0,0 +1,109 @@
+import type { FallbackEntry } from "./model-requirement-types.js"
+import type { FallbackModelObject } from "./fallback-model-object.js"
+import { normalizeModel } from "./model-normalization.js"
+import { resolveModelPipeline } from "./model-resolution-pipeline.js"
+import { KNOWN_VARIANTS } from "./known-variants.js"
+import type { ConnectedProvidersAdapter } from "./connected-providers-cache.js"
+import * as connectedProvidersCache from "./connected-providers-cache.js"
+
+export type ModelResolutionInput = {
+  userModel?: string
+  inheritedModel?: string
+  systemDefault?: string
+}
+
+export type ModelSource =
+  | "override"
+  | "category-default"
+  | "provider-fallback"
+  | "system-default"
+
+export type ModelResolutionResult = {
+  model: string
+  source: ModelSource
+  variant?: string
+}
+
+export type ExtendedModelResolutionInput = {
+  uiSelectedModel?: string
+  userModel?: string
+  userFallbackModels?: string[]
+  categoryDefaultModel?: string
+  fallbackChain?: FallbackEntry[]
+  availableModels: Set<string>
+  systemDefaultModel?: string
+}
+
+
+export function resolveModel(input: ModelResolutionInput): string | undefined {
+  return (
+    normalizeModel(input.userModel) ??
+    normalizeModel(input.inheritedModel) ??
+    input.systemDefault
+  )
+}
+
+export function resolveModelWithFallback(
+  input: ExtendedModelResolutionInput,
+  connectedProvidersAdapter: ConnectedProvidersAdapter = connectedProvidersCache,
+): ModelResolutionResult | undefined {
+  const { uiSelectedModel, userModel, userFallbackModels, categoryDefaultModel, fallbackChain, availableModels, systemDefaultModel } = input
+  const resolved = resolveModelPipeline({
+    intent: { uiSelectedModel, userModel, userFallbackModels, categoryDefaultModel },
+    constraints: { availableModels },
+    policy: { fallbackChain, systemDefaultModel },
+  }, connectedProvidersAdapter)
+
+  if (!resolved) {
+    return undefined
+  }
+
+  return {
+    model: resolved.model,
+    source: resolved.provenance,
+    variant: resolved.variant,
+  }
+}
+
+/**
+ * Normalizes fallback_models config to a mixed array.
+ * Accepts string, string[], or mixed arrays of strings and FallbackModelObject entries.
+ */
+export function normalizeFallbackModels(
+  models: string | (string | FallbackModelObject)[] | undefined,
+): (string | FallbackModelObject)[] | undefined {
+  if (!models) return undefined
+  if (typeof models === "string") return [models]
+  return models
+}
+
+/**
+ * Extracts plain model strings from a mixed fallback models array.
+ * Object entries are flattened to "model" or "model(variant)" strings.
+ * Use this when consumers need string[] (e.g., resolveModelForDelegateTask).
+ */
+export function flattenToFallbackModelStrings(
+  models: (string | FallbackModelObject)[] | undefined,
+): string[] | undefined {
+  if (!models) return undefined
+  return models.map((entry) => {
+    if (typeof entry === "string") return entry
+    const variant = entry.variant
+    if (variant) {
+      // Strip any supported inline variant syntax before appending explicit override.
+      // Supports both parenthesized and space-suffix forms so we don't emit
+      // invalid strings like "provider/model high(low)".
+      const model = entry.model
+        .replace(/\([^()]+\)\s*$/, "")
+        .replace(/\s+([a-z][a-z0-9_-]*)\s*$/i, (_match: string, suffix: string) => {
+          const normalized = String(suffix).toLowerCase()
+          return KNOWN_VARIANTS.has(normalized)
+            ? ""
+            : _match
+        })
+        .trim()
+      return `${model}(${variant})`
+    }
+    return entry.model
+  })
+}

apps/coder/src/services/model-resolution/provider-cache.ts

@@ -0,0 +1,27 @@
+export interface ModelMetadata {
+  readonly id: string
+  readonly provider?: string
+  readonly context?: number
+  readonly output?: number
+  readonly name?: string
+  readonly variants?: Record<string, unknown>
+  readonly limit?: {
+    readonly context?: number
+    readonly input?: number
+    readonly output?: number
+  }
+  readonly modalities?: {
+    readonly input?: string[]
+    readonly output?: string[]
+  }
+  readonly capabilities?: Record<string, unknown>
+  readonly reasoning?: boolean
+  readonly temperature?: boolean
+  readonly tool_call?: boolean
+  readonly [key: string]: unknown
+}
+
+export interface ProviderCache {
+  readConnectedProvidersCache(): string[] | null
+  findProviderModelMetadata(providerID: string, modelID: string): ModelMetadata | undefined
+}

apps/coder/src/services/model-resolution/provider-model-id-transform.ts

@@ -0,0 +1,69 @@
+function inferSubProvider(model: string): string | undefined {
+  if (model.startsWith("claude-")) return "anthropic"
+  if (model.startsWith("gpt-")) return "openai"
+  if (model.startsWith("gemini-")) return "google"
+  if (model.startsWith("grok-")) return "xai"
+  if (model.startsWith("minimax-")) return "minimax"
+  if (model.startsWith("kimi-")) return "moonshotai"
+  if (model.startsWith("glm-")) return "zai"
+  return undefined
+}
+
+const CLAUDE_VERSION_DOT = /claude-(\w+)-(\d+)-(\d+)/g
+const GEMINI_31_PRO_PREVIEW = /gemini-3\.1-pro(?!-)/g
+const GEMINI_3_FLASH_PREVIEW = /gemini-3-flash(?!-)/g
+
+function claudeVersionDot(model: string): string {
+  return model.replace(CLAUDE_VERSION_DOT, "claude-$1-$2.$3")
+}
+
+function applyGatewayTransforms(model: string): string {
+  return claudeVersionDot(model).replace(
+    GEMINI_31_PRO_PREVIEW,
+    "gemini-3.1-pro-preview",
+  )
+}
+
+function transformModelForProviderUsingAnthropicBehavior(
+  provider: string,
+  model: string,
+): string {
+  if (provider === "vercel") {
+    const slashIndex = model.indexOf("/")
+    if (slashIndex !== -1) {
+      const subProvider = model.substring(0, slashIndex)
+      const subModel = model.substring(slashIndex + 1)
+      return `${subProvider}/${applyGatewayTransforms(subModel)}`
+    }
+    const subProvider = inferSubProvider(model)
+    if (subProvider) {
+      return `${subProvider}/${applyGatewayTransforms(model)}`
+    }
+    return model
+  }
+  if (provider === "github-copilot") {
+    return claudeVersionDot(model)
+      .replace(GEMINI_31_PRO_PREVIEW, "gemini-3.1-pro-preview")
+      .replace(GEMINI_3_FLASH_PREVIEW, "gemini-3-flash-preview")
+  }
+  if (provider === "google") {
+    return model
+      .replace(GEMINI_31_PRO_PREVIEW, "gemini-3.1-pro-preview")
+      .replace(GEMINI_3_FLASH_PREVIEW, "gemini-3-flash-preview")
+  }
+  if (provider === "anthropic") {
+    return model
+  }
+  return model
+}
+
+export function transformModelForProvider(provider: string, model: string): string {
+  return transformModelForProviderUsingAnthropicBehavior(provider, model)
+}
+
+export function transformModelForProviderDisplay(
+  provider: string,
+  model: string,
+): string {
+  return transformModelForProviderUsingAnthropicBehavior(provider, model)
+}

data/AGENTS.md

@@ -17,7 +17,7 @@ top_p: 0.95
 top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_middleware, get_routes, get_semantic_neighborhoods, get_symbol_info, git_status, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output]
 description: Reviews code for bugs, security issues, and maintainability. Read-only.
 ---
 You review code. Find real problems, not style nits.
@@ -56,7 +56,7 @@ top_p: 0.95
 top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [ask_user_input, find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_semantic_neighborhoods, get_symbol_info, git_status, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output, watch_changes]
 description: Diagnoses bugs from error messages, logs, or described symptoms.
 ---
 You diagnose bugs. Form a hypothesis, prove it with evidence from the code.
@@ -82,7 +82,7 @@ top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
 steps: 5
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_middleware, get_routes, get_semantic_neighborhoods, get_symbol_info, git_status, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output, watch_changes]
 description: Proposes refactors for clarity, deduplication, or decoupling. Read-only — outputs plans, not edits.
 ---
 You propose refactors. You do not apply them. The user applies via OpenCode or Claude Code.
@@ -125,7 +125,7 @@ top_k: 20
 min_p: 0.0
 presence_penalty: 1.5
 steps: 20
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_middleware, get_routes, get_semantic_neighborhoods, get_symbol_info, git_status, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output, watch_changes, web_fetch, web_search]
 description: Designs new features, modules, or architectural changes. Outputs a build plan.
 ---
 You design. You produce build plans, not code.
@@ -167,7 +167,7 @@ top_p: 0.95
 top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output]
 description: Audits code for security vulnerabilities. Read-only.
 ---
 You audit for security issues. Concrete findings only, no generic warnings.
@@ -212,7 +212,7 @@ top_p: 0.95
 top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
-tools: [view_file, list_dir, grep, find_files]
+tools: [find_files, get_codebase_overview, grep, list_dir, view_file]
 description: Builds prompts for OpenCode, Claude Code, or BooCode dispatch.
 ---
 You write prompts that another coding agent will execute. Your output is the prompt, not the work.
@@ -250,7 +250,7 @@ top_p: 0.95
 top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_middleware, get_routes, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, request_read_access, search_symbols, view_file, view_truncated_output, watch_changes]
 description: Discovers and maps unfamiliar codebases. Reads architecture, traces data flow, identifies key symbols.
 ---
 You map codebases. Start broad, then drill into specifics.
@@ -278,7 +278,7 @@ top_k: 20
 min_p: 0.0
 presence_penalty: 0.0
 steps: 10
-tools: [find_files, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_semantic_neighborhoods, get_symbol_info, grep, list_dir, search_symbols, view_file, watch_changes, request_read_access, view_truncated_output, ask_user_input, git_status, get_blast_radius, get_hot_files, get_middleware, get_routes]
+tools: [ask_user_input, find_files, get_blast_radius, get_codebase_overview, get_dependencies, get_file_analysis, get_framework_analysis, get_hot_files, get_middleware, get_routes, get_semantic_neighborhoods, get_symbol_info, git_status, grep, list_dir, request_read_access, search_symbols, view_file, watch_changes]
 description: Produces actionable step plans from requirements. Read-only — never modifies files.
 ---
 You produce actionable step plans. You do not modify files.

data/skills/boocode/self-healing/SKILL.md

@@ -0,0 +1,287 @@
+---
+name: self-healing
+description: "Active runtime recovery for coding agents: when something breaks mid-task, diagnose the root cause, write a fix, VERIFY by re-running the broken thing, then file a `HEAL-` entry to `.learnings/HEALS.md` with proof. Use whenever a command, test, build, or lint fails or exits non-zero; on missing tooling, dependency/lockfile mismatch, wrong runtime version, venv or permission errors, port conflicts, dirty git state, or a missing `.env`; when the agent needs a helper or one-off script that doesn't exist yet; when an external API, tool, or MCP errors or rate-limits; or when a test flakes. Search `HEALS.md` by `Pattern-Key` first — most heals are recurrences, so increment `Recurrence-Count` instead of duplicating. Verify is mandatory: mark `pending-verify` honestly if sandboxed, `abandoned` if the fix can't be made to work. Pairs with `self-improvement` (which promotes recurring heals to durable memory) but owns the verify-before-persist discipline self-improvement doesn't."
+---
+
+# Self-Healing
+
+Active runtime recovery for coding agents. When something breaks, run the loop: **diagnose → patch → verify → file**. Leave behind a reusable, verified artifact instead of a swept-under-the-rug failure.
+
+The premise mirrors [browser-use/browser-harness](https://github.com/browser-use/browser-harness): *the harness improves itself every run*. An agent that hits a gap doesn't fail — it writes the fix during execution, verifies it works, and files the durable artifact for future runs. Coding tasks deserve the same loop.
+
+## What this skill is for
+
+When a coding agent hits a wall mid-task, the default failure modes are:
+
+1. **Paper over it** — "let me try a different approach" — and lose the recovery
+2. **Pretend the fix worked** — without re-running the broken thing
+3. **Symptom-fix** — skip the test, swallow the error, retry until green
+
+All three turn a one-time failure into a recurrence. The next agent on the same project hits the same wall.
+
+This skill enforces one discipline: **verify before persist**. A patch isn't real until you've re-run the failing operation and watched it succeed. When it does, file the verified fix so the next run benefits.
+
+## Relationship to self-improvement
+
+These two skills are deliberately split. Run both — they feed each other but don't overlap.
+
+| Aspect      | `self-healing` (this skill)                                          | `self-improvement`                                            |
+| ----------- | -------------------------------------------------------------------- | ------------------------------------------------------------- |
+| **When**    | During execution, failure is live                                    | After the fact, at natural breakpoints                        |
+| **Verb**    | Heal now — restore working state                                     | Remember for later — accumulate knowledge                     |
+| **Outcome** | Verified patch + (optional) reusable artifact                        | Logged learning, correction, request                          |
+| **Verify**  | **Mandatory** — no persist without proof                             | Not required                                                  |
+| **Files**   | `.learnings/HEALS.md` + `.learnings/heals/<HEAL-ID>/` (lazy)         | `.learnings/ERRORS.md`, `LEARNINGS.md`, `FEATURE_REQUESTS.md` |
+| **Trigger** | Failure observed mid-task                                            | Correction, knowledge gap, feature request, recurrence        |
+
+**Boundary rule:** if you're capturing a fact, a correction, or a wish — that's `self-improvement`. If you're applying and verifying a fix to a live failure — that's `self-healing`.
+
+## The Heal Loop
+
+```
+  ● failure observed
+  │
+  ● 1. DIAGNOSE  capture context — command, error, env, what was attempted
+  │              search HEALS.md for the same Pattern-Key first
+  │              (most heals are recurrences; don't reinvent)
+  │
+  ● 2. PATCH     write the fix — script, helper, env tweak, alt command
+  │              artifacts → .learnings/heals/<HEAL-ID>/  (only if needed)
+  │
+  ● 3. VERIFY    re-run the failing op — must succeed
+  │              ↻ if still failing: refine and retry, cap at 3 attempts
+  │              ✗ if uncrackable: file Status: abandoned with notes
+  │
+  ● 4. FILE      write HEAL-YYYYMMDD-XXX to .learnings/HEALS.md
+  │              with Pattern-Key, status, verification proof
+  │
+  ✓ working state restored, heal persisted
+
+  (conditional) PROMOTE  if Pattern-Key recurrence ≥ 3 across distinct tasks,
+                           append a Handoff block → self-improvement promotes to memory
+```
+
+If you abandon a heal mid-loop, don't pretend it succeeded. File a `HEAL-` entry with `Status: abandoned` and notes on what didn't work. The next agent learns from the dead end too.
+
+## When to trigger
+
+Self-healing fires on **active failures during execution** — the agent has just observed something not working and needs to make it work to continue. Five shapes:
+
+### 1. Tool failure (command / test / build / lint)
+Any invocation exits non-zero or produces wrong output. Don't acknowledge and retry verbatim — diagnose, patch, verify.
+
+*Examples:* `npm install` errors when a `pnpm-lock.yaml` is present (switch tool); `pytest` fails with `ModuleNotFoundError` (activate the venv); `tsc` flags a stale type (regenerate the client); `eslint` reports a config error (install the missing parser).
+
+### 2. Missing capability / tool gap
+The agent needs something that doesn't exist yet — a script, a helper, a wrapper, a glue function. Write it in the moment. This is the closest analog to browser-harness's `agent_helpers.py`.
+
+*Examples:* dedupe a CSV by custom key (write a small Python helper); bootstrap 12 microservices the same way (write `scripts/bootstrap-all.sh`); bulk-rename branches matching a pattern (write a `gh`-based shell helper).
+
+### 3. Environment issue
+The local environment isn't what the project expects. Detect, patch, verify.
+
+*Examples:* runtime version mismatch (`nvm use`, `pyenv local`, `rustup override`); stale dependency cache after a branch switch; dirty git state blocking a checkout; missing `.env` (copy from `.env.example` and surface gaps).
+
+### 4. External service / API change
+A service the agent depends on returns something unexpected. Find a workaround and capture it.
+
+*Examples:* an MCP tool returns `InputValidationError` because the schema changed (patch the call shape); a public API hits a rate limit (back off, switch endpoint, batch); an upstream lib bumped a default and broke a script (pin the version).
+
+### 5. About-to-retry-the-same-broken-approach
+The agent catches itself about to redo the failing step. That self-recognition is a heal forming — capture the alternate approach as the patch.
+
+### Detection signals to watch for
+
+- Non-zero exit codes
+- Stack traces in tool output
+- The same operation failing twice with the same error
+- "I'll try a different approach" — capture it as a heal
+- `command not found` / `module not found` / `permission denied`
+- Stale assertions, snapshot mismatches, type errors that weren't there before
+- "Weird" output that suggests environmental rather than logical bugs
+
+## HEAL Entry Format
+
+Append to `.learnings/HEALS.md` (create if missing):
+
+```markdown
+## [HEAL-YYYYMMDD-XXX] short_kebab_name
+
+**Logged**: ISO-8601 timestamp
+**Status**: verified | pending-verify | abandoned
+**Trigger**: tool-failure | missing-capability | env-issue | external-change | <free-form>
+**Active-Context**: (optional) — current skill, task phase, or workflow stage; omit if not applicable
+**Area**: free-form tag — what part of the system (`build`, `tests`, `ci`, `auth`, `data-pipeline`, `mobile`, ...)
+**Priority**: low | medium | high | critical
+
+### Failure
+What broke — concrete: the command, the error message, the action that was blocked. Include exit codes and verbatim error lines.
+
+### Diagnosis
+The root cause as understood after investigation. Why the obvious approach didn't work. Not a guess — what was actually verified during the heal.
+
+### Fix
+The patch that was applied. Verbatim commands, code snippets, or pointers to files under `.learnings/heals/<HEAL-ID>/`. Keep it minimal — just enough to reproduce.
+
+### Verification
+What was run after the fix and what it returned. Exit code, output snippet, test pass count. **This is the proof.** Without it, the entry is `pending-verify` or `abandoned`.
+
+### Artifacts
+(omit this section if no files were generated; otherwise list relative paths under `.learnings/heals/<HEAL-ID>/`)
+
+### Metadata
+- Related Files: path/to/file.ext
+- See Also: HEAL-... | LRN-... | ERR-... (related entries)
+- Pattern-Key: lower.snake.case key for recurrence detection (e.g. `env.lockfile_mismatch`)
+- Recurrence-Count: 1
+- First-Seen / Last-Seen: YYYY-MM-DD
+
+---
+```
+
+### Field guidance
+
+- **Status** — `verified` = the verify step passed. `pending-verify` = patch applied but couldn't be fully proven (sandboxed/offline/CI-only) — surface to the user. `abandoned` = patch didn't work or diagnosis was wrong — document what was tried.
+- **Trigger** — free-form is fine. The listed values are common shapes; what matters is that the failure shape is described enough for future agents to match against.
+- **Active-Context** — optional. Use it if your environment has a meaningful "what was I doing" tag (an active skill, a current task phase, a build stage, an agent role). Skip if not applicable. The browser-harness analog is the per-domain scoping of `domain-skills/<site>/`.
+- **Area** — free-form. Pick whatever helps future agents find this. `frontend`, `data-pipeline`, `ci`, `auth`, `terraform`, `mobile`, `embedded` — anything that fits your project shape.
+- **Pattern-Key** — lower.snake.case, stable, reusable across projects. Two heals with the same key are recurrences. `env.lockfile_mismatch` is good; `fixed_thing_tuesday` isn't.
+
+## ID generation
+
+Format: `HEAL-YYYYMMDD-XXX`. `XXX` is sequential 3-digit or 3-char random alphanumeric. Examples: `HEAL-20260524-001`, `HEAL-20260524-A7B`.
+
+## Artifacts directory (lazy)
+
+Only create `.learnings/heals/<HEAL-ID>/` when the heal generated something worth preserving. One-line fixes don't need a folder; the HEAL entry text is enough. Abandoned heals with no applied patch also skip the folder.
+
+```
+.learnings/
+├── HEALS.md
+├── ERRORS.md / LEARNINGS.md / FEATURE_REQUESTS.md  (self-improvement)
+└── heals/
+    └── HEAL-20260524-001/
+        ├── helper.sh
+        ├── patch.diff
+        └── notes.md
+```
+
+**Put here:** generated scripts/helpers, patch files, supplementary notes, output captures that document the diagnosis.
+**Don't put here:** project source changes (those go in the project tree, referenced via Related Files); secrets; output already captured in the HEAL text.
+
+## Verification rules
+
+Verify is the load-bearing wall. The whole point of self-healing over self-improvement is that the fix is *proven*, not theorized.
+
+### What counts as proof
+
+| Failure shape                         | Verification                                                       |
+| ------------------------------------- | ------------------------------------------------------------------ |
+| Tool / command / test / build / lint  | Re-run the original invocation; expect exit 0 / pass               |
+| Missing capability                    | Invoke the helper end-to-end on a real input; expect the intent    |
+| Environment drift                     | Re-run the operation that triggered the diagnosis                  |
+| External service workaround           | Re-run the failed call with the patch; expect a usable response    |
+
+### Sandboxed / offline / CI-only failures
+
+When you genuinely can't run the verify step (no network, no real remote, sandboxed shell, CI-only reproduction), file `Status: pending-verify` with:
+
+- The exact command the user / CI should run
+- The acceptance criteria — what counts as proof
+- A simulated proof if you can construct one (e.g. a dry-run mode, a stub of the failing call, a sandbox script)
+
+`pending-verify` is honest. Faking `verified` is the failure mode this skill exists to prevent.
+
+### When to invest in a proof script
+
+Most heals don't need a separate proof script — the verify step is just re-running the failing thing. Build a proper proof script when:
+
+- The heal generates a reusable helper that needs to be exercised across cases
+- The failure can't be reproduced live but can be reproduced in a sandbox (clean git repo, mock service, fake input)
+- You expect the heal to be re-applied across projects — the proof script then doubles as a regression check
+
+### If verification fails
+
+1. **Once** — refine the patch and retry. First diagnosis is often wrong.
+2. **Twice** — step back and reconsider the diagnosis. Maybe the root cause is elsewhere.
+3. **Three times** — stop. File `Status: abandoned` with notes on what you tried. Surface to the user. Don't flail.
+
+### What does NOT count as verification
+
+- "It looks right" / "I think this should work"
+- Re-running a *different* command than the one that originally failed
+- Suppressing the failure (`|| true`, `--ignore-errors`) — that's hiding
+- Skipping or deleting the failing test — that's regression
+- Passing because the cache was warm from before the fix
+
+### Reversibility
+
+Prefer reversible patches. If your heal modifies project files, capture the diff in `patch.diff`. If the heal is destructive (deletes generated files, rewrites locks), note it explicitly — a future agent reading the HEAL needs to know what was destroyed.
+
+## Recurrence and promotion
+
+Most heals are recurrences. Before filing a new HEAL, search:
+
+```bash
+grep -n "Pattern-Key: <your-pattern-key>" .learnings/HEALS.md
+```
+
+If found:
+
+- Increment `Recurrence-Count`
+- Update `Last-Seen`
+- Add the current occurrence as a See Also link
+- **Do not** create a duplicate entry
+
+### Promotion threshold
+
+Add a `Handoff` block to an existing entry when **all** are true:
+
+- `Recurrence-Count >= 3`
+- Seen across at least 2 distinct tasks
+- The fix is generalizable (not project-specific in a way that's already in a memory file)
+
+```markdown
+### Handoff
+- **Promoted To**: self-improvement at YYYY-MM-DD
+- **Promotion Target**: CLAUDE.md | AGENTS.md | .github/copilot-instructions.md | new-skill
+- **Distilled Rule**: One-line prevention guidance derived from the heal
+```
+
+Then `self-improvement` (or a learning aggregator) takes over: distills the rule, writes it into the right context file, or extracts a reusable skill. The HEAL stays for traceability.
+
+## Anti-patterns
+
+1. **Logging without verifying.** A HEAL filed before the fix is proven turns this into noisier self-improvement. If verify hasn't passed, the entry is `pending-verify` or `abandoned`.
+2. **Healing the symptom, not the cause.** A failing test isn't healed by skipping it (`pytest.skip`, `it.skip`, `xit`). A flaky CI isn't healed by `--retry`. Find the root cause; if you can't, abandon honestly.
+3. **Generating a new fix without trying existing ones first.** Search `HEALS.md` by Pattern-Key. Most heals are recurrences.
+4. **Inventing helpers when the project already has them.** Look in `scripts/`, `Makefile`, `justfile`, `package.json`, `pyproject.toml` first. Heal = write what's missing, not what's there.
+5. **Scope creep.** A heal is scoped to one failure. Cleanup belongs in a quality pass; refactors are features. Scope creep makes heals unreviewable.
+6. **Empty artifact folders.** Don't create `.learnings/heals/<HEAL-ID>/` if nothing goes in it.
+
+## Best practices
+
+1. **Heal eagerly, file always.** Even abandoned heals teach the next agent what doesn't work.
+2. **Verify before persist.** The non-negotiable rule.
+3. **Minimal and reversible patches.** A 3-line fix is a heal; a 300-line refactor is a feature.
+4. **Stable Pattern-Keys.** `env.node_version_mismatch` is reusable; `fixed_the_thing_on_tuesday` isn't.
+5. **Reference, don't duplicate.** Cross-link related HEAL/LRN/ERR via See Also.
+6. **Hand off recurrences.** A heal seen 3 times deserves to be in the project's permanent memory.
+7. **Don't gate the main tree on heal artifacts.** Files under `.learnings/heals/` are reference material; if a script becomes load-bearing, promote it to `scripts/`.
+
+## Setup
+
+```bash
+mkdir -p .learnings        # heals/ is lazy — created only when artifacts exist
+touch .learnings/HEALS.md
+```
+
+Gitignore choices match `self-improvement`. Keep heals local (`.learnings/` in `.gitignore`) or share them as team knowledge (don't gitignore — they become reviewable durable context).
+
+## Multi-agent use
+
+The skill is agent-agnostic. The `.learnings/HEALS.md` format is plain markdown — any agent (Claude Code, BooCode agents, OpenCode, Copilot, Cursor, Aider, ...) can read and write it.
+
+## See also
+
+- [`references/examples.md`](references/examples.md) — canonical HEAL entry shapes (command failure, missing capability, env drift, external API workaround, abandoned heal)

data/skills/boocode/self-healing/eval.yaml

@@ -0,0 +1,35 @@
+skill: self-healing
+tasks:
+  - prompt: "I'm in a project root that has pnpm-lock.yaml present but no package-lock.json. I just tried to run `npm install` and it failed. Get me to a working state so I can keep working — I have other things to do, just unblock me. After fixing it, make sure future agents in this project know what happened."
+    grader:
+      - the response invokes the self-healing skill
+      - the response diagnoses pnpm vs npm mismatch as the root cause
+      - the response runs pnpm install successfully
+      - the response files a HEAL entry to .learnings/HEALS.md with Status: verified
+      - the HEAL entry has Trigger: tool-failure
+      - the HEAL entry has a Pattern-Key resembling env.lockfile_mismatch
+      - the HEAL entry includes the verification output
+  - prompt: "I need to bulk-rename 8 git branches in this repo from `feat-XXX-name` to `feat/XXX-name`. There's no existing script for this and `gh` doesn't have a bulk-rename. Write what's needed, prove it works on a dry run, and capture the work so it's not lost if I need it again."
+    grader:
+      - the response invokes the self-healing skill
+      - the response recognizes this as a missing-capability heal
+      - the response writes a helper script under .learnings/heals/HEAL-<date>-<seq>/
+      - the response runs a dry-run verification
+      - the response files a HEAL entry with Status: verified and Trigger: missing-capability
+      - the HEAL entry references the helper script in the Artifacts section
+  - prompt: "I just ran `pytest` and got `ModuleNotFoundError: No module named 'pydantic'`. There's already a `.learnings/HEALS.md` in this project with a prior heal for a similar venv-not-activated issue. Fix this, and do the right thing with the heal records."
+    grader:
+      - the response invokes the self-healing skill
+      - the response searches HEALS.md first (using find-similar-heals.sh or grep) before writing a new fix
+      - the response finds the existing HEAL entry and applies its fix (activate venv)
+      - the response increments Recurrence-Count on the existing entry
+      - the response updates Last-Seen on the existing entry
+      - the response does NOT create a duplicate HEAL entry
+  - prompt: "A test in this repo is failing intermittently — the snapshot for `Card.test.tsx` flakes. I've already tried fixing it once by stubbing the date; it passes twice then flakes again because there's a UUID that's also non-deterministic. I don't have time to refactor the Card component to inject dependencies. Just do the right thing — get me to a state that's honest about what's known and not known, and don't pretend the heal worked."
+    grader:
+      - the response invokes the self-healing skill
+      - the response diagnoses that the initial patch attempt was incomplete
+      - the response files a HEAL entry with Status: abandoned
+      - the HEAL entry documents what was tried and why it failed
+      - the response does NOT mark anything as verified
+      - the response surfaces the situation honestly to the user

data/skills/boocode/self-healing/references/examples.md

@@ -0,0 +1,248 @@
+# Self-Healing Examples
+
+Concrete HEAL entries showing the format applied to real failure shapes. Use these as templates when filing your own heals. All examples use the iteration-2 schema (free-form `Trigger` / `Area`, optional `Active-Context`, no `Source` field, lazy artifact folders).
+
+---
+
+## Example 1 — Tool failure (lockfile mismatch)
+
+```markdown
+## [HEAL-20260524-001] npm_install_pnpm_lockfile
+
+**Logged**: 2026-05-24T14:22:01Z
+**Status**: verified
+**Trigger**: tool-failure
+**Area**: build
+**Priority**: medium
+
+### Failure
+`npm install` exited 1 with `npm ERR! code EUSAGE` and a notice that `pnpm-lock.yaml` is present but `package-lock.json` is missing. The project uses pnpm workspaces; npm refuses to install against a pnpm lockfile.
+
+### Diagnosis
+Project root contains `pnpm-lock.yaml`. The README and CI both invoke `pnpm`. `npm` was a habit from previous projects, not the actual project's package manager.
+
+### Fix
+Use pnpm instead:
+```bash
+pnpm install
+```
+
+### Verification
+```
+$ pnpm install
+Lockfile is up to date, resolution step is skipped
+Already up to date
+✓ Done in 1.4s
+```
+Exit 0.
+
+### Metadata
+- Related Files: package.json, pnpm-lock.yaml
+- See Also: (none yet)
+- Pattern-Key: env.lockfile_mismatch
+- Recurrence-Count: 1
+- First-Seen: 2026-05-24
+- Last-Seen: 2026-05-24
+
+---
+```
+
+Pattern-Key `env.lockfile_mismatch` is reusable across projects (yarn.lock, bun.lockb, etc.). At Recurrence ≥ 3, this should be promoted to `CLAUDE.md` or `AGENTS.md` as a verification step.
+
+No Artifacts section — the fix is a tool swap, no files generated. Lazy folder pattern: nothing to put in `.learnings/heals/HEAL-20260524-001/`, so the folder isn't created.
+
+---
+
+## Example 2 — Missing capability (helper written on the fly)
+
+```markdown
+## [HEAL-20260524-002] bulk_rename_branches_helper
+
+**Logged**: 2026-05-24T15:10:44Z
+**Status**: verified
+**Trigger**: missing-capability
+**Area**: ci
+**Priority**: low
+
+### Failure
+Need to rename 12 feature branches from `feat-XXX-name` to `feat/XXX-name`. No existing project script handles this; `gh` doesn't have a bulk-rename primitive.
+
+### Diagnosis
+This is glue work, not a project bug. A small shell helper using `gh api` per branch is the right level — not worth a top-level script, but worth keeping the file for the next time someone asks.
+
+### Fix
+Wrote `.learnings/heals/HEAL-20260524-002/rename-branches.sh`:
+
+```bash
+#!/usr/bin/env bash
+set -euo pipefail
+git fetch --all
+for branch in $(git branch -r | grep 'origin/feat-' | sed 's|origin/||'); do
+  new="${branch/feat-/feat/}"
+  echo "$branch → $new"
+  gh api -X POST "repos/{owner}/{repo}/git/refs" \
+    -f "ref=refs/heads/$new" \
+    -f "sha=$(git rev-parse "origin/$branch")"
+  gh api -X DELETE "repos/{owner}/{repo}/git/refs/heads/$branch"
+done
+```
+
+### Verification
+Dry-run (commented out the API calls) printed the 12 expected mappings.
+Live run renamed all 12; `git branch -r | grep 'feat-' | wc -l` returns 0.
+
+### Artifacts
+- `.learnings/heals/HEAL-20260524-002/rename-branches.sh`
+
+### Metadata
+- Related Files: (none — operates on git refs)
+- See Also: (none)
+- Pattern-Key: tool.gh.bulk_branch_rename
+- Recurrence-Count: 1
+- First-Seen: 2026-05-24
+- Last-Seen: 2026-05-24
+
+---
+```
+
+Helper script lives under `.learnings/heals/<HEAL-ID>/` — referenceable, but not assumed to be load-bearing. If it gets reused frequently, promote to `scripts/`.
+
+---
+
+## Example 3 — Environment issue (runtime version)
+
+```markdown
+## [HEAL-20260524-003] nvm_use_project_node
+
+**Logged**: 2026-05-24T16:01:12Z
+**Status**: verified
+**Trigger**: env-issue
+**Active-Context**: verify-gate
+**Area**: tests
+**Priority**: medium
+
+### Failure
+`pnpm test` exited 1 with `engine "node" is incompatible with this module. Expected version "^20.10.0". Got "18.19.0"`.
+
+### Diagnosis
+`.nvmrc` requests node 20.10.0; current shell has 18.19.0 from a previous project context. The shell's nvm wasn't switched after `cd`-ing into the repo.
+
+### Fix
+```bash
+nvm use   # reads .nvmrc
+```
+
+### Verification
+```
+$ node --version
+v20.10.0
+$ pnpm test
+✓ 47 tests passed
+```
+
+### Metadata
+- Related Files: .nvmrc, package.json
+- See Also: (none)
+- Pattern-Key: env.node_version_mismatch
+- Recurrence-Count: 1
+- First-Seen: 2026-05-24
+- Last-Seen: 2026-05-24
+
+---
+```
+
+`Active-Context: verify-gate` because that's the workflow phase the agent was in when the test step blew up. An upstream context loader could surface this entry next time `verify-gate` runs in a node project. If you don't have an analogous concept in your pipeline, omit the field.
+
+---
+
+## Example 4 — External service workaround
+
+```markdown
+## [HEAL-20260524-004] gh_api_rate_limit_backoff
+
+**Logged**: 2026-05-24T17:33:08Z
+**Status**: verified
+**Trigger**: external-change
+**Area**: ci
+**Priority**: high
+
+### Failure
+Looping `gh api repos/.../issues` over 200 issues started returning `403 rate limit exceeded` after ~60 calls. Unauthenticated burst limit (abuse detection on rapid successive calls).
+
+### Diagnosis
+Script was using `gh api` REST without batching. `gh` is authenticated but the secondary rate limit fires on rapid successive calls — not the primary 5000/hour limit. Switching to a single paginated GraphQL query bypasses the secondary limit entirely.
+
+### Fix
+```bash
+gh api graphql -f query='
+  query($owner:String!,$repo:String!,$cursor:String) {
+    repository(owner:$owner,name:$repo) {
+      issues(first:100,after:$cursor) { ... }
+    }
+  }' -F owner=... -F repo=...
+```
+Took ~3 calls total instead of 200.
+
+### Verification
+Full run completed in 4.8s, no 403s, all 200 issues retrieved. Compared output against a sample of the original per-issue calls — fields match.
+
+### Artifacts
+- `.learnings/heals/HEAL-20260524-004/fetch-issues.sh`
+
+### Metadata
+- Related Files: (none — ad-hoc query)
+- See Also: (none)
+- Pattern-Key: api.gh.rate_limit
+- Recurrence-Count: 1
+- First-Seen: 2026-05-24
+- Last-Seen: 2026-05-24
+
+---
+```
+
+---
+
+## Example 5 — Abandoned heal (diagnosis was wrong)
+
+```markdown
+## [HEAL-20260524-005] vitest_flaky_snapshot
+
+**Logged**: 2026-05-24T18:14:22Z
+**Status**: abandoned
+**Trigger**: tool-failure
+**Active-Context**: verify-gate
+**Area**: tests
+**Priority**: medium
+
+### Failure
+`vitest` snapshot test `Card > renders default` flaked twice in three runs. Diff showed a timestamp string differing by ~3 seconds.
+
+### Diagnosis (initial — wrong)
+Assumed flake was timezone drift in the snapshot fixture. Patched the fixture to use a fixed `Date.now()` stub.
+
+### Diagnosis (current — correct)
+The snapshot depends on multiple non-deterministic values: timestamp AND a `crypto.randomUUID()`. The clock stub addressed only one of them. The UUID is still random per render, so the snapshot keeps drifting on subsequent runs.
+
+### Fix (attempted)
+Added `vi.useFakeTimers({ now: 1700000000000 })` to the test setup.
+
+### Verification
+Test passed twice, then flaked again on the third run — same `Card > renders default`, different diff (this time the UUID changed). Original diagnosis was incomplete.
+
+### Abandonment notes
+The right fix is to make the component deterministic via dependency injection (pass a `clock` and `idGen` prop), not to stub globally. That's a real change to the component contract — out of scope for a heal. Filed `FEAT-20260524-001` via self-improvement; surfaced to the user.
+
+### Metadata
+- Related Files: src/components/Card.tsx, src/components/Card.test.tsx
+- See Also: FEAT-20260524-001
+- Pattern-Key: tests.flaky_snapshot_multi_nondeterminism
+- Recurrence-Count: 1
+- First-Seen: 2026-05-24
+- Last-Seen: 2026-05-24
+
+---
+```
+
+Abandoned heals are first-class. They document a dead end so the next agent doesn't re-walk it. The handoff to a `FEAT-` entry via self-improvement is the right next step when the real fix is a feature, not a heal.
+
+No Artifacts section — the attempted patch was reverted; nothing reusable was generated.

data/skills/boocode/self-healing/scripts/detect-failure.sh

@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+# detect-failure.sh — PostToolUse hook for Bash invocations.
+# Reads the tool result JSON on stdin (per Claude Code hook spec); if exit_code != 0,
+# emits a system reminder pointing the agent at self-healing.
+#
+# Wire up in .claude/settings.json:
+#   "hooks": {
+#     "PostToolUse": [{ "matcher": "Bash",
+#       "hooks": [{ "type": "command",
+#         "command": "./data/skills/boocode/self-healing/scripts/detect-failure.sh" }] }]
+#   }
+
+set -euo pipefail
+
+# Hook payload arrives on stdin. We tolerate either jq-style JSON or raw text.
+PAYLOAD="$(cat || true)"
+
+# Try to parse exit_code; fall through silently on parse failure.
+EXIT_CODE=$(printf '%s' "$PAYLOAD" | python3 -c '
+import json, sys
+try:
+    data = json.loads(sys.stdin.read() or "{}")
+    # Common shapes: {"tool_result": {"exit_code": N}}, {"exit_code": N}, {"output": "...", "exit_code": N}
+    for path in (("tool_result","exit_code"), ("exit_code",), ("result","exit_code")):
+        d = data
+        ok = True
+        for k in path:
+            if isinstance(d, dict) and k in d:
+                d = d[k]
+            else:
+                ok = False
+                break
+        if ok and isinstance(d, int):
+            print(d)
+            sys.exit(0)
+except Exception:
+    pass
+print(0)
+' 2>/dev/null || echo 0)
+
+if [[ "$EXIT_CODE" != "0" ]]; then
+  cat <<'EOF'
+<self-healing-trigger>
+A Bash command just exited non-zero. This is a heal opportunity.
+
+Before retrying the same command verbatim:
+  1. DIAGNOSE — read the error; identify the root cause (env? missing dep? wrong tool?)
+  2. Search .learnings/HEALS.md for a matching Pattern-Key (don't re-solve a solved problem)
+  3. PATCH — write the fix (or apply a known one)
+  4. VERIFY — re-run the command; require exit 0
+  5. FILE — append a HEAL entry to .learnings/HEALS.md via data/skills/boocode/self-healing/scripts/new-heal.sh
+</self-healing-trigger>
+EOF
+fi

data/skills/boocode/self-healing/scripts/find-similar-heals.sh

@@ -0,0 +1,52 @@
+#!/usr/bin/env bash
+# find-similar-heals.sh — Search existing heals before generating a new fix.
+# Usage: ./find-similar-heals.sh <pattern-key-or-keyword>
+#
+# Prints matching HEAL entries with their Pattern-Key, Status, and Recurrence-Count
+# so the agent can decide whether to re-apply an existing fix or write a new one.
+
+set -euo pipefail
+
+QUERY="${1:-}"
+HEALS_FILE="$(pwd)/.learnings/HEALS.md"
+
+if [[ -z "$QUERY" ]]; then
+  echo "usage: $0 <pattern-key-or-keyword>" >&2
+  exit 2
+fi
+
+if [[ ! -f "$HEALS_FILE" ]]; then
+  echo "(no .learnings/HEALS.md yet — no prior heals to consult)"
+  exit 0
+fi
+
+# Find HEAL section headers that contain the query in their body (Pattern-Key, name, or text).
+python3 - <<PY "$QUERY" "$HEALS_FILE"
+import sys, re
+query, path = sys.argv[1].lower(), sys.argv[2]
+with open(path) as f:
+    text = f.read()
+# Split into entries by ^## [HEAL-...]
+entries = re.split(r"(?m)^## \[HEAL-", text)[1:]
+hits = []
+for body in entries:
+    if query in body.lower():
+        head = body.splitlines()[0]
+        pk = re.search(r"Pattern-Key:\s*(\S+)", body)
+        status = re.search(r"Status\*\*:\s*(\S+)", body) or re.search(r"Status:\s*(\S+)", body)
+        rc = re.search(r"Recurrence-Count:\s*(\d+)", body)
+        hits.append({
+            "id": "HEAL-" + head.split("]")[0],
+            "name": head.split("]", 1)[1].strip() if "]" in head else head,
+            "pattern_key": pk.group(1) if pk else "?",
+            "status": status.group(1) if status else "?",
+            "recurrence": rc.group(1) if rc else "1",
+        })
+if not hits:
+    print(f"(no heals match '{query}')")
+else:
+    print(f"Found {len(hits)} matching heal(s):\n")
+    for h in hits:
+        print(f"  {h['id']} {h['name']}")
+        print(f"    pattern={h['pattern_key']}  status={h['status']}  recurrence={h['recurrence']}")
+PY

data/skills/boocode/self-healing/scripts/new-heal.sh

@@ -0,0 +1,74 @@
+#!/usr/bin/env bash
+# new-heal.sh — Initialize a new HEAL-<date>-<seq> entry skeleton.
+# Usage: ./new-heal.sh <short_kebab_name> [trigger]
+#   trigger: tool-failure | missing-capability | env-issue | external-change | <free-form>
+#
+# Appends a templated HEAL entry to .learnings/HEALS.md and prints the HEAL-ID.
+# Does NOT create .learnings/heals/<HEAL-ID>/ — that folder is lazy, created
+# only when artifacts are written.
+
+set -euo pipefail
+
+NAME="${1:-}"
+TRIGGER="${2:-tool-failure}"
+
+if [[ -z "$NAME" ]]; then
+  echo "usage: $0 <short_kebab_name> [trigger]" >&2
+  exit 2
+fi
+
+LEARNINGS_DIR="$(pwd)/.learnings"
+HEALS_FILE="$LEARNINGS_DIR/HEALS.md"
+mkdir -p "$LEARNINGS_DIR"
+
+DATE="$(date +%Y%m%d)"
+SEQ=$(grep -c "^## \[HEAL-${DATE}-" "$HEALS_FILE" 2>/dev/null || echo 0)
+NEXT=$(printf "%03d" $((SEQ + 1)))
+HEAL_ID="HEAL-${DATE}-${NEXT}"
+
+# Active-Context is optional. The agent / harness can set ACTIVE_CONTEXT in env.
+ACTIVE_CONTEXT="${ACTIVE_CONTEXT:-}"
+ACTIVE_LINE=""
+if [[ -n "$ACTIVE_CONTEXT" ]]; then
+  ACTIVE_LINE="**Active-Context**: $ACTIVE_CONTEXT
+"
+fi
+
+cat >> "$HEALS_FILE" <<EOF
+
+## [$HEAL_ID] $NAME
+
+**Logged**: $(date -u +%Y-%m-%dT%H:%M:%SZ)
+**Status**: pending-verify
+**Trigger**: $TRIGGER
+${ACTIVE_LINE}**Area**: TODO
+**Priority**: medium
+
+### Failure
+TODO — concrete error, command, exit code
+
+### Diagnosis
+TODO — root cause after investigation
+
+### Fix
+TODO — patch applied (commands, snippets, or pointers to .learnings/heals/$HEAL_ID/ if files were generated)
+
+### Verification
+TODO — what was run after the fix, what it returned. **Update Status to "verified" only after this passes.**
+
+### Metadata
+- Related Files: TODO
+- See Also: TODO
+- Pattern-Key: TODO
+- Recurrence-Count: 1
+- First-Seen: $(date +%Y-%m-%d)
+- Last-Seen: $(date +%Y-%m-%d)
+
+---
+EOF
+
+# stdout = the HEAL-ID alone, so `ID=$(new-heal.sh ...)` captures it cleanly.
+# Human guidance goes to stderr.
+echo "$HEAL_ID"
+echo "$HEALS_FILE" >&2
+echo "(create .learnings/heals/$HEAL_ID/ only if you generate artifacts to put there)" >&2

data/skills/boocode/verify-gate/SKILL.md

@@ -0,0 +1,178 @@
+---
+name: verify-gate
+description: "Runs project compile, test, and lint commands between implementation and quality review. Gates simplify-and-harden behind machine verification. If checks fail, enters a fix loop with diagnostics. If checks pass, signals ready for quality pass. Use after any implementation work completes and before signaling done. Essential for the inner loop's verify step."
+---
+
+# Verify Gate
+
+Machine verification gate between implementation and quality review. Runs the project's compile, test, and lint commands. If any fail, enters a fix loop. If all pass, unblocks the quality pass.
+
+This is the inner loop's **verify** step. Without it, the agent hands off code with zero machine signal about whether it actually works.
+
+## When to Use
+
+- After any implementation work completes, before signaling "done"
+- Before running simplify-and-harden or quality review
+- After fixing audit findings from code review
+- Any time you want a machine-verified green signal
+
+## Pipeline Position
+
+```
+[implementation] → verify-gate → [quality review / simplify-and-harden]
+                   ↻ fix loop — on failure, diagnose and retry
+```
+
+## Step 1: Discover Project Commands
+
+Read the project's configuration to find verification commands. Check these sources in order:
+
+1. **Project instruction files** (`CLAUDE.md`, `data/AGENTS.md`) — look for a `## Verification` or `## Test Commands` section
+2. **package.json** — `scripts.test`, `scripts.lint`, `scripts.typecheck`, `scripts.build`. BooCode uses pnpm, so prefer `pnpm run <script>` when `pnpm-lock.yaml` is present.
+3. **Makefile** / **Justfile** — `test`, `lint`, `check`, `build` targets
+4. **Cargo.toml** — `cargo build`, `cargo test`, `cargo clippy`
+5. **pyproject.toml** / **setup.cfg** — `pytest`, `mypy`, `ruff`
+6. **go.mod** — `go build ./...`, `go test ./...`, `go vet ./...`
+7. **deno.json** / **deno.jsonc** — `deno task <name>` for any defined tasks
+
+If no commands are discoverable, ask the user once and suggest they add a `## Verification` section to `CLAUDE.md` for future sessions:
+
+```markdown
+## Verification
+
+- Build: `pnpm run build`
+- Test: `pnpm test`
+- Lint: `pnpm run lint`
+- Type check: `npx tsc -p apps/server/tsconfig.json --noEmit`
+```
+
+## Step 2: Run Verification
+
+Run discovered commands in this order. Stop at the first failure category.
+
+### Phase 1: Compile / Type Check
+Run the build or type-check command. These catch structural errors before wasting time on tests.
+
+```
+Exit 0 → proceed to Phase 2
+Exit non-zero → enter fix loop with compiler output
+```
+
+### Phase 2: Tests
+Run the test command. Scope to changed files if the test runner supports it.
+
+```
+Exit 0 → proceed to Phase 3
+Exit non-zero → enter fix loop with test output
+```
+
+### Phase 3: Lint (optional, skippable with --skip-lint)
+Run the lint command. Lint failures are lower severity but still worth catching.
+
+```
+Exit 0 → all phases green, gate passes
+Exit non-zero → enter fix loop with lint output
+```
+
+## Step 3: Fix Loop
+
+When a phase fails:
+
+1. **Read the output.** Parse the error output for actionable diagnostics — file paths, line numbers, error messages.
+2. **Scope the fix.** Only fix what the verification caught. Do not refactor, improve, or touch unrelated code.
+3. **Apply the fix.** Make the minimal change to resolve the failure.
+4. **Re-run the failed phase.** Not all phases — just the one that failed.
+5. **If it passes**, continue to the next phase.
+6. **If it fails again**, increment the attempt counter.
+
+### Fix Loop Limits
+
+- **Default max attempts:** 3 per phase (configurable via `--fix-limit N`)
+- **Counter increments on every attempt**, even if the error changes. Fixing Error A and uncovering Error B counts as attempt 2, not attempt 1. The counter tracks fix attempts, not unique errors.
+- **If limit reached:** Stop. Report what failed, what was tried, and the remaining error output. Do not guess further — signal to the user that manual intervention is needed.
+- **Total budget:** The fix loop should not exceed 20% of the original implementation effort. If fixes are snowballing, stop and report.
+
+## Step 4: Gate Signal
+
+When all phases pass:
+
+```markdown
+## Verify Gate: PASSED
+
+- Build: passed
+- Tests: passed (N tests, M suites)
+- Lint: passed (or skipped)
+
+Ready for quality review.
+```
+
+When the fix loop is exhausted:
+
+```markdown
+## Verify Gate: BLOCKED
+
+- Build: passed
+- Tests: FAILED (attempt 3/3)
+  - [file:line] error description
+  - [file:line] error description
+- Lint: not reached
+
+Fix loop exhausted. Manual intervention needed before quality review.
+```
+
+## Integration with Other Skills
+
+### boocode simplify-and-harden / quality review
+verify-gate should gate any quality pass. Run verify-gate first; only proceed to review if the gate passes.
+
+### self-healing (if available)
+On any failure during the verify run, consider handing the diagnostics to a self-healing loop (diagnose → patch → verify → persist). Verify-gate then re-runs the checks. Up to 3 heal attempts per phase before abandoning.
+
+### self-improvement
+If a recurring error pattern emerges across verify runs, capture it in `CLAUDE.md` or as a new skill under `data/skills/boocode/` so future verify-gate runs don't rediscover the same fix.
+
+## What This Skill Does NOT Do
+
+- Does not review code quality (that's a separate review pass)
+- Does not check security
+- Does not verify spec compliance
+- Does not modify test files or add new tests
+- Does not run tests for code it didn't change (unless the test runner doesn't support scoping)
+
+## Configuration
+
+If the project has a `verify-gate` section in `CLAUDE.md` or `data/AGENTS.md`:
+
+```yaml
+## Verify Gate Config
+
+build: pnpm run build
+test: pnpm test
+lint: pnpm run lint
+type_check: npx tsc -p apps/server/tsconfig.json --noEmit
+fix_limit: 3
+skip_lint: false
+test_scope: changed  # changed | all
+```
+
+If no configuration exists, discover commands automatically (Step 1) and suggest persisting them.
+
+### Custom Verification Steps
+
+Projects with custom invariants can define additional verification phases. These run as extra phases after the standard compile/test/lint checks.
+
+Example — a project that needs API schema validation:
+
+```yaml
+## Verify Gate Config
+
+custom_checks:
+  - name: validate-schema
+    command: python scripts/validate_schema.py --strict
+  - name: check-no-legacy-imports
+    command: grep -r "from legacy" src/ --include="*.py" && exit 1 || exit 0
+```
+
+When custom checks are defined, verify-gate runs them as **Phase 4** after lint. Each check's exit code determines pass/fail. Failed checks enter the same fix loop as standard phases.
+
+This moves project-specific invariants from "knowledge in your head" to "knowledge in the harness" — exactly where the agent can reach it.