- Parallel batch execution: batch field on Step, batchConfig on Flow,
batch-aware readySteps with maxConcurrent gating, getReadyInBatch helper
- SWITCH branching step: new 'switch' StepKind with cases/programmed conditions,
resolveSwitch() pure function, switch-excluded steps tracked in
SchedulerState, non-selected branches excluded from execution
- Add 'timed_out' to flow_runs/flow_steps CHECK constraints
- Add retry_count and max_retries columns to flow_steps
- Add timeout detection in advanceInner loop (configurable FLOW_STEP_TIMEOUT_MS)
- Add retriable logic: re-dispatch on timeout if maxRetries > 0 and retryCount < maxRetries
- Add isRetriable() + shouldRetry() pure decision functions
- Add timed_out handling to reconcileResumeStep and reconcileRun
- Add 'timed_out' to ws-frames enum, publishStep status type
- Remove codecontext service block from docker-compose.yml
- Remove CODECONTEXT_URL env var
- Delete codecontext/Dockerfile
- Update callCodecontext() to try boocontext MCP first with HTTP fallback
- Graceful degradation: if boocontext MCP unavailable, tools still work via HTTP
New plans table (id, project_id, title, description, status, flow_run_id,
progress_pct, items_total, items_completed, metadata, timestamps) with
CHECK constraints and indexes.
Plan store (plan-store.ts): createPlan, getPlan, listPlans, listActivePlans,
updatePlan, updatePlanFromRun, findPlanWithRunningRun, planStatusFromRun.
Flow-runner integration: onRunTerminal callback fires on every terminal
transition (complete/fail/cancel) and updates linked plans automatically.
5 API endpoints: GET /api/plans, GET /api/plans/active, GET /api/plans/:id,
POST /api/plans, PATCH /api/plans/:id.
484 tests pass, build clean.
Bridge script that calls paseo import <session-id> --provider opencode
--label omo=true on task() child sessions. Supports import, archive,
ls commands with --dry-run verification. Skill at .opencode/skills/
is gitignored (user-level) — copy from scripts/ on setup.
Shared boocontext MCP client (boocontext_client.ts) wrapping the existing
mcp-client.ts callTool() infrastructure with 32KB truncation and error
handling. Used by get_code_health.
4 new first-class agent tools backed by the boocontext MCP server:
- get_code_health — A-F grades per file across 7 dimensions, project health
summary, refactoring candidates (wraps boocontext_health)
- get_code_impact — merged symbol trace + blast radius in one call (wraps
boocontext_impact, replaces two-step get_symbol_info+get_blast_radius)
- get_type_info — TypeScript type recovery via type-inject MCP (wraps
boocontext_types, returns signatures, interfaces, generics, JSDoc)
- get_code_map — DCP-compressed context map with compress toggle (wraps
boocontext_map, 10x token reduction vs full scan)
All 4 registered in ALL_TOOLS as read-only tools.
In-memory SessionMeta registry tracks active terminal sessions with
paneId, sessionId, projectPath, title, createdAt, lastActivityAt.
GET /api/term/sessions returns all active sessions as JSON array.
Registry is updated on WS attach and cleaned up on disconnect.
Registers @nick-vi/type-inject-mcp as a stdio MCP server via npx.
Provides lookup_type and list_types tools for TypeScript type
recovery — solves the 0% TS type recovery gap in codecontext.
KV cache quantization (--cache-type-k q4_0) and ngram speculative decoding
(--spec-type ngram-mod) are high-value llama.cpp features that improve VRAM
usage and tokens/sec. Removing them from the shadowing lists allows agents
to enable them via llama_extra_args.
New @boocode/ion package (v0.0.1) for inference optimization network.
.codesight/ wiki artifacts for codebase documentation.
.omo/ work plans for openspec cleanup and enhanced file panel.
codecontext Dockerfile and docker-compose adjustments for sidecar build.
MCP example config cleanup (remove deprecated entries). pnpm-lock.yaml
updated for new dependencies.
Implements audit-harness-inspired session lifecycle: audit session
creation/end/recover/report-daily with JSONL buffer and graded context
recovery (L0-L4). Guideline service for behavioral compliance rules
(condition/action model with criticality). Correction service for
persistent user correction tracking across agent sessions.
8 supporting skills: audit-start/end/report-daily/recover + command
variants for slash-command integration.
Adds Inference tab to SettingsPane with controls for temperature, top-p,
top-k, min-p, and other inference parameters. Server-side route and
provider config wiring to pass overrides through the inference pipeline.
New /analytics route: token usage dashboard with aggregate summary,
per-session breakdown, context window stats, and per-category token
distribution. Data served from existing agent_sessions + tool_cost_stats.
New /results route: browsable archive of orchestrator flow runs and
arena battles. Two-tab layout (Analysis Runs / Arena Battles) using
existing API endpoints (no new backend).
Sidebar gains Results (ScrollText icon) and Token Analytics (BarChart3
icon) nav buttons above Settings.
Drop 9 batch proposals that are superseded by the boocode-lift-analysis
(boocontext-audit, conductor upgrades, self-healing/verify-gate skills):
add-3tier-memory, import-llm-evaluator, import-pregel-engine, plugin-platform,
conductor-evolution, code-intelligence-upgrade, dev-workflow, ui-overhaul,
agent-reliability.
Delete 11 stub archive files (49-66B each, 'Status: Shipped. Archived.' only)
that provide zero documentation value over the existing CHANGELOG.md + git tags.
- Approval gate steps pause and await human resolution
- appendStepEvent wired into markStep, failRun, dispatchAgentStep
- Trigger rule unit tests (6 variants)
- New parallel-research flow with one_success trigger
- TriggerRule type (all_success/one_success/all_done) for parallel deps
- Variable substitution ($stepId.output.field) in agent step prompts
- Approval gate step kind (pauses flow via permission frames)
- flow_step_events table for append-only event-sourced step log
- evaluateTriggerRule pure function in flow-runner-decisions
