indifferentketchup/boocode
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

v1.12.4: complete inference.ts split into services/inference/

Browse Source 
- sentinel-summaries.ts: runCapHitSummary, insertCapHitSentinel,
  runDoomLoopSummary, insertDoomLoopSentinel
- inference.ts → inference/turn.ts: residue is runAssistantTurn,
  runInference, createInferenceRunner orchestration only
- inference/index.ts: re-export shim preserves the public surface
  (createInferenceRunner, runInference, runAssistantTurn,
  detectDoomLoop, DOOM_LOOP_THRESHOLD, buildMessagesPayload, plus
  type-side InferenceContext/InferenceFrame/StreamResult/TurnArgs/
  FramePublisher)
- src/index.ts + auto_name.ts + the two vitest test files updated to
  import from ./services/inference/index.js explicitly (NodeNext ESM
  doesn't honor directory-index resolution)

Final tally: 11 files under services/inference/, the largest being
sentinel-summaries.ts at 523 LoC (two near-clone summary paths kept
side-by-side until a third sentinel justifies factoring out a shared
runWrapUpSummary). turn.ts is now 326 LoC, the next-largest is
stream-phase.ts at 380. Public import surface unchanged.

tool-phase.ts → turn.ts back-edge for runAssistantTurn remains
(cycle is safe; resolved at call time).

Prepares the file structure for v1.13 AI SDK migration — streamText
swap targets stream-phase.ts only.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
indifferentketchup
2026-05-21 22:36:35 +00:00
parent c87df6981a
commit 9ef00c0268
11 changed files with 368 additions and 327 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
apps/server/src/index.ts
View File

@@ -16,7 +16,7 @@ import { registerWebSocket } from './routes/ws.js';
import { registerModelRoutes } from './routes/models.js'; import { registerModelRoutes } from './routes/models.js';
import { registerAgentRoutes } from './routes/agents.js'; import { registerAgentRoutes } from './routes/agents.js';
import { registerSkillsRoutes } from './routes/skills.js'; import { registerSkillsRoutes } from './routes/skills.js';
import { createInferenceRunner } from './services/inference.js'; import { createInferenceRunner } from './services/inference/index.js';
import { createBroker } from './services/broker.js'; import { createBroker } from './services/broker.js';
import { listSkills } from './services/skills.js'; import { listSkills } from './services/skills.js';
import * as compaction from './services/compaction.js'; import * as compaction from './services/compaction.js';

2
apps/server/src/services/__tests__/doom-loop.test.ts
View File

@@ -1,5 +1,5 @@
import { describe, it, expect } from 'vitest'; import { describe, it, expect } from 'vitest';
import { DOOM_LOOP_THRESHOLD, detectDoomLoop } from '../inference.js'; import { DOOM_LOOP_THRESHOLD, detectDoomLoop } from '../inference/index.js';
import type { ToolCall } from '../../types/api.js'; import type { ToolCall } from '../../types/api.js';
// ---- fixture ---------------------------------------------------------------- // ---- fixture ----------------------------------------------------------------

2
apps/server/src/services/__tests__/inference.test.ts
View File

@@ -1,5 +1,5 @@
import { describe, it, expect } from 'vitest'; import { describe, it, expect } from 'vitest';
import { buildMessagesPayload } from '../inference.js'; import { buildMessagesPayload } from '../inference/index.js';
import type { import type {
Message, Message,
MessageRole, MessageRole,

2
apps/server/src/services/auto_name.ts
View File

@@ -1,4 +1,4 @@
import type { InferenceContext } from './inference.js'; import type { InferenceContext } from './inference/index.js';
const NAMING_SYSTEM_PROMPT = const NAMING_SYSTEM_PROMPT =
'You name chat sessions. Reply directly with no thinking, reasoning, or explanation. Output ONLY the title, 4 words max, no quotes, no punctuation, no prefix like "Title:".'; 'You name chat sessions. Reply directly with no thinking, reasoning, or explanation. Output ONLY the title, 4 words max, no quotes, no punctuation, no prefix like "Title:".';

2
apps/server/src/services/inference/error-handler.ts
View File

@@ -1,7 +1,7 @@
import type { MessageMetadata, Session } from '../../types/api.js'; import type { MessageMetadata, Session } from '../../types/api.js';
import * as modelContext from '../model-context.js'; import * as modelContext from '../model-context.js';
import { maybeFlagForCompaction } from './payload.js'; import { maybeFlagForCompaction } from './payload.js';
import type { InferenceContext, StreamResult, TurnArgs } from '../inference.js'; import type { InferenceContext, StreamResult, TurnArgs } from './turn.js';
export async function handleAbortOrError( export async function handleAbortOrError(
ctx: InferenceContext, ctx: InferenceContext,

20
apps/server/src/services/inference/index.ts Normal file
View File

@@ -0,0 +1,20 @@
// v1.12.4: re-export shim. Outside callers (apps/server/src/index.ts and the
// vitest inference tests) import from './services/inference/index.js'. The
// directory is now the public surface; turn.ts holds runAssistantTurn /
// runInference / createInferenceRunner while the other inference/*.ts files
// stay implementation-private.
export {
createInferenceRunner,
runAssistantTurn,
runInference,
} from './turn.js';
export type {
FramePublisher,
InferenceContext,
InferenceFrame,
StreamResult,
TurnArgs,
} from './turn.js';
export { detectDoomLoop, DOOM_LOOP_THRESHOLD } from './sentinels.js';
export { buildMessagesPayload } from './payload.js';

2
apps/server/src/services/inference/payload.ts
View File

@@ -8,7 +8,7 @@ import type {
import * as compaction from '../compaction.js'; import * as compaction from '../compaction.js';
import { buildSystemPrompt } from '../system-prompt.js'; import { buildSystemPrompt } from '../system-prompt.js';
import { isAnySentinel } from './sentinels.js'; import { isAnySentinel } from './sentinels.js';
import type { InferenceContext } from '../inference.js'; import type { InferenceContext } from './turn.js';
export interface OpenAiMessage { export interface OpenAiMessage {
role: 'system' | 'user' | 'assistant' | 'tool'; role: 'system' | 'user' | 'assistant' | 'tool';

331
apps/server/src/services/inference.ts → apps/server/src/services/inference/sentinel-summaries.ts
View File

@@ -1,47 +1,20 @@
import type { FastifyBaseLogger } from 'fastify';
import type { Sql } from '../db.js';
import type { Config } from '../config.js';
import type { import type {
Agent, Agent,
ErrorReason,
Message, Message,
MessageMetadata, MessageMetadata,
Project, Project,
Session, Session,
ToolCall, } from '../../types/api.js';
UserStreamFrame, import * as modelContext from '../model-context.js';
} from '../types/api.js'; import { buildMessagesPayload } from './payload.js';
import { ALL_TOOLS } from './tools.js'; import { DOOM_LOOP_THRESHOLD } from './sentinels.js';
import { resolveProjectRoot } from './path_guard.js'; import { streamCompletion } from './stream-phase.js';
import { maybeAutoNameChat } from './auto_name.js'; import { DB_FLUSH_INTERVAL_MS } from './types.js';
import { getAgentById } from './agents.js'; import type {
import * as compaction from './compaction.js'; InferenceContext,
import * as modelContext from './model-context.js'; StreamResult,
import type { Broker } from './broker.js'; TurnArgs,
import { resolveToolBudget } from './inference/budget.js'; } from './turn.js';
import {
DOOM_LOOP_THRESHOLD,
detectDoomLoop,
} from './inference/sentinels.js';
import {
buildMessagesPayload,
loadContext,
} from './inference/payload.js';
import {
finalizeCompletion,
handleAbortOrError,
} from './inference/error-handler.js';
import {
executeStreamPhase,
streamCompletion,
} from './inference/stream-phase.js';
import { executeToolPhase } from './inference/tool-phase.js';
import { DB_FLUSH_INTERVAL_MS, type StreamPhaseState } from './inference/types.js';
// v1.12.4: re-exported so external callers (tests, future consumers) keep
// importing from services/inference.js as the public surface.
export { detectDoomLoop, DOOM_LOOP_THRESHOLD } from './inference/sentinels.js';
export { buildMessagesPayload } from './inference/payload.js';
// Synthetic system note appended to the cap-hit summary call. Verbatim from // Synthetic system note appended to the cap-hit summary call. Verbatim from
// the v1.8.2 spec — do not paraphrase: the model is more reliable when the // the v1.8.2 spec — do not paraphrase: the model is more reliable when the
@@ -52,219 +25,7 @@ const CAP_HIT_SUMMARY_NOTE = (limit: number) =>
const DOOM_LOOP_NOTE = (name: string) => const DOOM_LOOP_NOTE = (name: string) =>
`You called ${name} with the same arguments ${DOOM_LOOP_THRESHOLD} times in a row. Stop calling it. Produce the best answer you can with what you have.`; `You called ${name} with the same arguments ${DOOM_LOOP_THRESHOLD} times in a row. Stop calling it. Produce the best answer you can with what you have.`;
export interface InferenceFrame { export async function runCapHitSummary(
type:
| 'message_started'
| 'delta'
| 'tool_call'
| 'tool_result'
| 'message_complete'
| 'usage'
| 'messages_deleted'
| 'session_renamed'
| 'chat_renamed'
| 'error';
message_id?: string;
message_ids?: string[];
chat_id?: string;
tool_message_id?: string;
tool_call_id?: string;
// v1.8.2: 'system' added so cap-hit sentinel messages can announce themselves
// through the normal message_started → delta → message_complete sequence.
role?: 'assistant' | 'tool' | 'user' | 'system';
content?: string;
tool_call?: ToolCall;
output?: unknown;
truncated?: boolean;
error?: string;
// v1.8.2: structured error reason. Set on `type: 'error'` so the UI can
// surface a specific message; `error` stays the human-readable text.
reason?: ErrorReason;
// v1.8.2: piggybacks on `message_complete` so static or terminally-resolved
// messages can carry their persisted metadata to the live stream without a
// refetch (sentinels carry { kind: 'cap_hit', ... }; failed messages carry
// { kind: 'error', ... }).
metadata?: MessageMetadata | null;
tokens_used?: number | null;
ctx_used?: number | null;
ctx_max?: number | null;
completion_tokens?: number | null;
started_at?: string | null;
finished_at?: string | null;
model?: string;
session_id?: string;
name?: string;
}
export type FramePublisher = (sessionId: string, frame: InferenceFrame) => void;
export interface InferenceContext {
sql: Sql;
config: Config;
log: FastifyBaseLogger;
publish: FramePublisher;
publishUser: (frame: UserStreamFrame) => void;
// v1.11: passed through so compaction.process can publish 'compacted'
// frames on the same session WS channel useSessionStream subscribes to.
// Compaction is the only path that needs the raw broker handle (regular
// inference goes through `publish`); keeping a separate field avoids
// tempting other code paths into bypassing the session-id binding.
broker: Broker;
}
// v1.12.4: payload assembly extracted to ./inference/payload.ts (tests
// import buildMessagesPayload from this module, so a re-export below
// preserves the public surface). Stream + tool phases extracted to
// ./inference/stream-phase.ts and ./inference/tool-phase.ts.
export interface StreamResult {
finishReason: string | null;
content: string;
toolCalls: ToolCall[];
promptTokens: number | null;
completionTokens: number | null;
}
export interface TurnArgs {
sessionId: string;
chatId: string;
assistantMessageId: string;
// v1.8.2: cumulative tool calls executed this run. Compared against the
// resolved budget at the top of each turn. Replaces the older `depth`
// counter (which counted iterations, not invocations).
toolsUsed: number;
// v1.11.6: ordered tool calls executed in this user-message turn (across
// recursive runAssistantTurn invocations). Reset to [] at user-message
// boundaries by runInference, same as toolsUsed. Doom-loop check at the
// top of runAssistantTurn slices the last DOOM_LOOP_THRESHOLD entries.
recentToolCalls: ToolCall[];
signal: AbortSignal | undefined;
}
export async function runAssistantTurn(
ctx: InferenceContext,
args: TurnArgs,
): Promise<void> {
const { sessionId, chatId } = args;
// v1.11: if the prior turn flagged this chat for compaction, run it first
// so loadContext below reads the post-compaction history. We swallow
// compaction failures (clearing the flag so we don't loop) and proceed
// with the un-compacted history — a slow turn that hits the model's
// hard limit is recoverable; a dead session is not.
const chatFlag = await ctx.sql<{ needs_compaction: boolean }[]>`
SELECT needs_compaction FROM chats WHERE id = ${chatId}
`;
if (chatFlag[0]?.needs_compaction) {
try {
await compaction.process({
sql: ctx.sql,
config: ctx.config,
log: ctx.log,
broker: ctx.broker,
chatId,
});
} catch (err) {
ctx.log.warn({ err, chatId }, 'auto-compaction failed; clearing flag and proceeding');
await ctx.sql`UPDATE chats SET needs_compaction = false WHERE id = ${chatId}`;
}
}
const loaded = await loadContext(ctx.sql, sessionId, chatId);
if (!loaded) {
ctx.log.warn({ sessionId }, 'inference: session or project missing');
return;
}
const { session, project, history } = loaded;
const projectRoot = await resolveProjectRoot(project.path);
// Agent resolution is per-turn so PATCH agent_id mid-conversation takes
// effect on the next message. Unknown agent_id returns null silently —
// session falls back to base prompt + all tools + default temperature.
const agent = session.agent_id
? await getAgentById(project.path, session.agent_id)
: null;
// v1.8.2: cap-hit replaces the older "tool loop depth exceeded" failure.
// When we've already burned the budget *before* this turn even runs, we
// skip straight to the summary flow — the in-flight assistant message slot
// gets reused for the wrap-up reply instead of being marked failed.
const budget = resolveToolBudget(agent);
if (args.toolsUsed >= budget) {
await runCapHitSummary(ctx, args, session, project, history, agent, budget);
return;
}
// v1.11.6: doom-loop guard. Detected BEFORE the budget cap (the model can
// burn through 3 identical calls long before the 15-call budget fires).
// Same in-flight-slot-reuse pattern as runCapHitSummary — wrap-up reply
// lands in args.assistantMessageId, then a doom_loop sentinel is inserted
// to make the abort visible in the chat history.
const loop = detectDoomLoop(args.recentToolCalls);
if (loop) {
await runDoomLoopSummary(ctx, args, session, project, history, agent, loop);
return;
}
const messages = await buildMessagesPayload(session, project, history, agent);
// v1.11.8: resolve per-chat web-tools opt-in. Tri-state on the wire:
// - session.web_search_enabled = null → inherit project default
// - session.web_search_enabled = true/false → explicit
// Both web_search and web_fetch are gated by this single flag (the UI
// label is "Enable web search and fetch" — same store, both tools).
// Default is false unless explicitly opted in, matching the v1.9
// plumbing intent ("inert until Batch 8 ships the actual tools").
const webToolsEnabled =
session.web_search_enabled ?? project.default_web_search_enabled ?? false;
const state: StreamPhaseState = { accumulated: '', startedAt: null };
let result: StreamResult;
try {
result = await executeStreamPhase(ctx, args, session, messages, state, agent, webToolsEnabled);
} catch (err) {
await handleAbortOrError(ctx, args, state.accumulated, err);
return;
}
if (result.toolCalls.length > 0) {
await executeToolPhase(ctx, args, result, state.startedAt, session, projectRoot);
return;
}
await finalizeCompletion(ctx, args, result, state.startedAt, session);
}
export async function runInference(
ctx: InferenceContext,
sessionId: string,
chatId: string,
assistantMessageId: string,
signal?: AbortSignal
): Promise<void> {
// v1.8.2: every fresh inference (initial send, regenerate, force_send,
// continue) starts with a clean budget. Tool-call accumulation across
// Continue invocations is what the hard ceiling guards against, not the
// per-call budget.
// v1.11.6: recentToolCalls also resets — doom-loop detection is scoped
// to a single user-message turn, so a Continue starts with no history.
return runAssistantTurn(ctx, {
sessionId,
chatId,
assistantMessageId,
toolsUsed: 0,
recentToolCalls: [],
signal,
});
}
// v1.8.2: cap-hit summary flow. Called instead of erroring when the loop
// hits its budget. Reuses the in-flight assistant message slot to stream a
// short wrap-up reply with the synthetic note prepended and tools disabled,
// then always inserts a cap_hit sentinel afterward (regardless of summary
// outcome) so the UI can show a Continue affordance.
async function runCapHitSummary(
ctx: InferenceContext, ctx: InferenceContext,
args: TurnArgs, args: TurnArgs,
session: Session, session: Session,
@@ -526,7 +287,7 @@ async function insertCapHitSentinel(
// Kept as a clone rather than refactored into a shared helper because the // Kept as a clone rather than refactored into a shared helper because the
// two summary paths still differ in error reason + sentinel shape; a third // two summary paths still differ in error reason + sentinel shape; a third
// sentinel would justify factoring out runWrapUpSummary(opts). // sentinel would justify factoring out runWrapUpSummary(opts).
async function runDoomLoopSummary( export async function runDoomLoopSummary(
ctx: InferenceContext, ctx: InferenceContext,
args: TurnArgs, args: TurnArgs,
session: Session, session: Session,
@@ -760,69 +521,3 @@ async function insertDoomLoopSentinel(
metadata, metadata,
}); });
} }
interface InferenceRegistration {
controller: AbortController;
completed: Promise<void>;
}
export function createInferenceRunner(
ctx: Omit<InferenceContext, 'publishUser'>,
publishUserFn: (user: string, frame: UserStreamFrame) => void
) {
const registry = new Map<string, InferenceRegistration>();
return {
enqueue(sessionId: string, chatId: string, assistantMessageId: string, user: string) {
const callCtx: InferenceContext = {
...ctx,
publishUser: (frame) => publishUserFn(user, frame),
// v1.11: broker comes in via ctx (set at registration time). Repeated
// here so the destructure carries it onto the per-call ctx without
// having to add it to every enqueue/cancel signature individually.
broker: ctx.broker,
};
// v1.8 mobile-tabs: announce working before the async loop starts so
// every device subscribed to the user channel sees the amber dot.
callCtx.publishUser({ type: 'chat_status', chat_id: chatId, status: 'streaming', at: new Date().toISOString() });
const controller = new AbortController();
let resolveCompleted!: () => void;
const completed = new Promise<void>((res) => { resolveCompleted = res; });
const registration: InferenceRegistration = { controller, completed };
registry.set(chatId, registration);
void (async () => {
try {
await runInference(callCtx, sessionId, chatId, assistantMessageId, controller.signal);
setImmediate(() => {
void maybeAutoNameChat(callCtx, chatId, sessionId).catch((err: Error) => {
callCtx.log.warn({ err, chatId }, 'auto-name failed');
});
});
} catch (err) {
callCtx.log.error({ err }, 'unhandled inference error');
} finally {
resolveCompleted();
// Only clear our own registration; a force-send may have replaced it.
if (registry.get(chatId) === registration) {
registry.delete(chatId);
}
}
})();
},
async cancel(_sessionId: string, chatId: string): Promise<boolean> {
const reg = registry.get(chatId);
if (!reg) return false;
reg.controller.abort();
// Swallow — we just need to wait for the catch/finally to persist state.
await reg.completed.catch(() => {});
return true;
},
hasActive(chatId: string): boolean {
return registry.has(chatId);
},
};
}
export const _toolNames = ALL_TOOLS.map((t) => t.name);

2
apps/server/src/services/inference/stream-phase.ts
View File

@@ -17,7 +17,7 @@ import type {
InferenceContext, InferenceContext,
StreamResult, StreamResult,
TurnArgs, TurnArgs,
} from '../inference.js'; } from './turn.js';
interface ChatCompletionDelta { interface ChatCompletionDelta {
role?: string; role?: string;

4
apps/server/src/services/inference/tool-phase.ts
View File

@@ -7,12 +7,12 @@ import type {
InferenceContext, InferenceContext,
StreamResult, StreamResult,
TurnArgs, TurnArgs,
} from '../inference.js'; } from './turn.js';
// v1.12.4: ESM value-import cycle. executeToolPhase recurses into // v1.12.4: ESM value-import cycle. executeToolPhase recurses into
// runAssistantTurn which lives in inference.ts. The cycle is safe because // runAssistantTurn which lives in inference.ts. The cycle is safe because
// the reference is read at call time (inside an async function body), not // the reference is read at call time (inside an async function body), not
// at module top-level. Node + tsc resolve this cleanly. // at module top-level. Node + tsc resolve this cleanly.
import { runAssistantTurn } from '../inference.js'; import { runAssistantTurn } from './turn.js';
async function executeToolCall( async function executeToolCall(
projectRoot: string, projectRoot: string,

326
apps/server/src/services/inference/turn.ts Normal file
View File

@@ -0,0 +1,326 @@
import type { FastifyBaseLogger } from 'fastify';
import type { Sql } from '../../db.js';
import type { Config } from '../../config.js';
import type {
Agent,
ErrorReason,
Message,
MessageMetadata,
Project,
Session,
ToolCall,
UserStreamFrame,
} from '../../types/api.js';
import { ALL_TOOLS } from '../tools.js';
import { resolveProjectRoot } from '../path_guard.js';
import { maybeAutoNameChat } from '../auto_name.js';
import { getAgentById } from '../agents.js';
import * as compaction from '../compaction.js';
import * as modelContext from '../model-context.js';
import type { Broker } from '../broker.js';
import { resolveToolBudget } from './budget.js';
import {
DOOM_LOOP_THRESHOLD,
detectDoomLoop,
} from './sentinels.js';
import {
buildMessagesPayload,
loadContext,
} from './payload.js';
import {
finalizeCompletion,
handleAbortOrError,
} from './error-handler.js';
import {
executeStreamPhase,
streamCompletion,
} from './stream-phase.js';
import { executeToolPhase } from './tool-phase.js';
import { DB_FLUSH_INTERVAL_MS, type StreamPhaseState } from './types.js';
import {
runCapHitSummary,
runDoomLoopSummary,
} from './sentinel-summaries.js';
// v1.12.4: re-exported so external callers (tests, future consumers) keep
// importing from services/inference.js as the public surface.
export { detectDoomLoop, DOOM_LOOP_THRESHOLD } from './sentinels.js';
export { buildMessagesPayload } from './payload.js';
export interface InferenceFrame {
type:
| 'message_started'
| 'delta'
| 'tool_call'
| 'tool_result'
| 'message_complete'
| 'usage'
| 'messages_deleted'
| 'session_renamed'
| 'chat_renamed'
| 'error';
message_id?: string;
message_ids?: string[];
chat_id?: string;
tool_message_id?: string;
tool_call_id?: string;
// v1.8.2: 'system' added so cap-hit sentinel messages can announce themselves
// through the normal message_started → delta → message_complete sequence.
role?: 'assistant' | 'tool' | 'user' | 'system';
content?: string;
tool_call?: ToolCall;
output?: unknown;
truncated?: boolean;
error?: string;
// v1.8.2: structured error reason. Set on `type: 'error'` so the UI can
// surface a specific message; `error` stays the human-readable text.
reason?: ErrorReason;
// v1.8.2: piggybacks on `message_complete` so static or terminally-resolved
// messages can carry their persisted metadata to the live stream without a
// refetch (sentinels carry { kind: 'cap_hit', ... }; failed messages carry
// { kind: 'error', ... }).
metadata?: MessageMetadata | null;
tokens_used?: number | null;
ctx_used?: number | null;
ctx_max?: number | null;
completion_tokens?: number | null;
started_at?: string | null;
finished_at?: string | null;
model?: string;
session_id?: string;
name?: string;
}
export type FramePublisher = (sessionId: string, frame: InferenceFrame) => void;
export interface InferenceContext {
sql: Sql;
config: Config;
log: FastifyBaseLogger;
publish: FramePublisher;
publishUser: (frame: UserStreamFrame) => void;
// v1.11: passed through so compaction.process can publish 'compacted'
// frames on the same session WS channel useSessionStream subscribes to.
// Compaction is the only path that needs the raw broker handle (regular
// inference goes through `publish`); keeping a separate field avoids
// tempting other code paths into bypassing the session-id binding.
broker: Broker;
}
// v1.12.4: payload assembly extracted to ./inference/payload.ts (tests
// import buildMessagesPayload from this module, so a re-export below
// preserves the public surface). Stream + tool phases extracted to
// ./inference/stream-phase.ts and ./inference/tool-phase.ts.
export interface StreamResult {
finishReason: string | null;
content: string;
toolCalls: ToolCall[];
promptTokens: number | null;
completionTokens: number | null;
}
export interface TurnArgs {
sessionId: string;
chatId: string;
assistantMessageId: string;
// v1.8.2: cumulative tool calls executed this run. Compared against the
// resolved budget at the top of each turn. Replaces the older `depth`
// counter (which counted iterations, not invocations).
toolsUsed: number;
// v1.11.6: ordered tool calls executed in this user-message turn (across
// recursive runAssistantTurn invocations). Reset to [] at user-message
// boundaries by runInference, same as toolsUsed. Doom-loop check at the
// top of runAssistantTurn slices the last DOOM_LOOP_THRESHOLD entries.
recentToolCalls: ToolCall[];
signal: AbortSignal | undefined;
}
export async function runAssistantTurn(
ctx: InferenceContext,
args: TurnArgs,
): Promise<void> {
const { sessionId, chatId } = args;
// v1.11: if the prior turn flagged this chat for compaction, run it first
// so loadContext below reads the post-compaction history. We swallow
// compaction failures (clearing the flag so we don't loop) and proceed
// with the un-compacted history — a slow turn that hits the model's
// hard limit is recoverable; a dead session is not.
const chatFlag = await ctx.sql<{ needs_compaction: boolean }[]>`
SELECT needs_compaction FROM chats WHERE id = ${chatId}
`;
if (chatFlag[0]?.needs_compaction) {
try {
await compaction.process({
sql: ctx.sql,
config: ctx.config,
log: ctx.log,
broker: ctx.broker,
chatId,
});
} catch (err) {
ctx.log.warn({ err, chatId }, 'auto-compaction failed; clearing flag and proceeding');
await ctx.sql`UPDATE chats SET needs_compaction = false WHERE id = ${chatId}`;
}
}
const loaded = await loadContext(ctx.sql, sessionId, chatId);
if (!loaded) {
ctx.log.warn({ sessionId }, 'inference: session or project missing');
return;
}
const { session, project, history } = loaded;
const projectRoot = await resolveProjectRoot(project.path);
// Agent resolution is per-turn so PATCH agent_id mid-conversation takes
// effect on the next message. Unknown agent_id returns null silently —
// session falls back to base prompt + all tools + default temperature.
const agent = session.agent_id
? await getAgentById(project.path, session.agent_id)
: null;
// v1.8.2: cap-hit replaces the older "tool loop depth exceeded" failure.
// When we've already burned the budget *before* this turn even runs, we
// skip straight to the summary flow — the in-flight assistant message slot
// gets reused for the wrap-up reply instead of being marked failed.
const budget = resolveToolBudget(agent);
if (args.toolsUsed >= budget) {
await runCapHitSummary(ctx, args, session, project, history, agent, budget);
return;
}
// v1.11.6: doom-loop guard. Detected BEFORE the budget cap (the model can
// burn through 3 identical calls long before the 15-call budget fires).
// Same in-flight-slot-reuse pattern as runCapHitSummary — wrap-up reply
// lands in args.assistantMessageId, then a doom_loop sentinel is inserted
// to make the abort visible in the chat history.
const loop = detectDoomLoop(args.recentToolCalls);
if (loop) {
await runDoomLoopSummary(ctx, args, session, project, history, agent, loop);
return;
}
const messages = await buildMessagesPayload(session, project, history, agent);
// v1.11.8: resolve per-chat web-tools opt-in. Tri-state on the wire:
// - session.web_search_enabled = null → inherit project default
// - session.web_search_enabled = true/false → explicit
// Both web_search and web_fetch are gated by this single flag (the UI
// label is "Enable web search and fetch" — same store, both tools).
// Default is false unless explicitly opted in, matching the v1.9
// plumbing intent ("inert until Batch 8 ships the actual tools").
const webToolsEnabled =
session.web_search_enabled ?? project.default_web_search_enabled ?? false;
const state: StreamPhaseState = { accumulated: '', startedAt: null };
let result: StreamResult;
try {
result = await executeStreamPhase(ctx, args, session, messages, state, agent, webToolsEnabled);
} catch (err) {
await handleAbortOrError(ctx, args, state.accumulated, err);
return;
}
if (result.toolCalls.length > 0) {
await executeToolPhase(ctx, args, result, state.startedAt, session, projectRoot);
return;
}
await finalizeCompletion(ctx, args, result, state.startedAt, session);
}
export async function runInference(
ctx: InferenceContext,
sessionId: string,
chatId: string,
assistantMessageId: string,
signal?: AbortSignal
): Promise<void> {
// v1.8.2: every fresh inference (initial send, regenerate, force_send,
// continue) starts with a clean budget. Tool-call accumulation across
// Continue invocations is what the hard ceiling guards against, not the
// per-call budget.
// v1.11.6: recentToolCalls also resets — doom-loop detection is scoped
// to a single user-message turn, so a Continue starts with no history.
return runAssistantTurn(ctx, {
sessionId,
chatId,
assistantMessageId,
toolsUsed: 0,
recentToolCalls: [],
signal,
});
}
// v1.8.2: cap-hit summary flow. Called instead of erroring when the loop
// hits its budget. Reuses the in-flight assistant message slot to stream a
// short wrap-up reply with the synthetic note prepended and tools disabled,
// then always inserts a cap_hit sentinel afterward (regardless of summary
// outcome) so the UI can show a Continue affordance.
interface InferenceRegistration {
controller: AbortController;
completed: Promise<void>;
}
export function createInferenceRunner(
ctx: Omit<InferenceContext, 'publishUser'>,
publishUserFn: (user: string, frame: UserStreamFrame) => void
) {
const registry = new Map<string, InferenceRegistration>();
return {
enqueue(sessionId: string, chatId: string, assistantMessageId: string, user: string) {
const callCtx: InferenceContext = {
...ctx,
publishUser: (frame) => publishUserFn(user, frame),
// v1.11: broker comes in via ctx (set at registration time). Repeated
// here so the destructure carries it onto the per-call ctx without
// having to add it to every enqueue/cancel signature individually.
broker: ctx.broker,
};
// v1.8 mobile-tabs: announce working before the async loop starts so
// every device subscribed to the user channel sees the amber dot.
callCtx.publishUser({ type: 'chat_status', chat_id: chatId, status: 'streaming', at: new Date().toISOString() });
const controller = new AbortController();
let resolveCompleted!: () => void;
const completed = new Promise<void>((res) => { resolveCompleted = res; });
const registration: InferenceRegistration = { controller, completed };
registry.set(chatId, registration);
void (async () => {
try {
await runInference(callCtx, sessionId, chatId, assistantMessageId, controller.signal);
setImmediate(() => {
void maybeAutoNameChat(callCtx, chatId, sessionId).catch((err: Error) => {
callCtx.log.warn({ err, chatId }, 'auto-name failed');
});
});
} catch (err) {
callCtx.log.error({ err }, 'unhandled inference error');
} finally {
resolveCompleted();
// Only clear our own registration; a force-send may have replaced it.
if (registry.get(chatId) === registration) {
registry.delete(chatId);
}
}
})();
},
async cancel(_sessionId: string, chatId: string): Promise<boolean> {
const reg = registry.get(chatId);
if (!reg) return false;
reg.controller.abort();
// Swallow — we just need to wait for the catch/finally to persist state.
await reg.completed.catch(() => {});
return true;
},
hasActive(chatId: string): boolean {
return registry.has(chatId);
},
};
}
export const _toolNames = ALL_TOOLS.map((t) => t.name);