maestro-flow 0.3.9 → 0.3.11

package/.codex/skills/team-testing/SKILL.md

@@ -1,235 +1,242 @@
----
-name: team-testing
-description: Unified team skill for testing team. Progressive test coverage through Generator-Critic loops, shared memory, and dynamic layer selection. Triggers on "team testing".
-allowed-tools: spawn_agent(*), wait_agent(*), send_message(*), followup_task(*), close_agent(*), list_agents(*), report_agent_job_result(*), request_user_input(*), Read(*), Write(*), Edit(*), Bash(*), Glob(*), Grep(*), mcp__maestro-tools__team_msg(*)
----
-
-# Team Testing
-
-Orchestrate multi-agent test pipeline: strategist -> generator -> executor -> analyst. Progressive layer coverage (L1/L2/L3) with Generator-Critic loops for coverage convergence.
-
-## Architecture
-
-```
-Skill(skill="team-testing", args="task description")
-                    |
-         SKILL.md (this file) = Router
-                    |
-     +--------------+--------------+
-     |                             |
-  no --role flag              --role <name>
-     |                             |
-  Coordinator                  Worker
-  roles/coordinator/role.md    roles/<name>/role.md
-     |
-     +-- analyze -> dispatch -> spawn workers -> STOP
-                                    |
-                    +-------+-------+-------+-------+
-                    v       v       v       v
-                [strat] [gen]  [exec]  [analyst]
-                team-worker agents, each loads roles/<role>/role.md
-```
-
-## Role Registry
-
-| Role | Path | Prefix | Inner Loop |
-|------|------|--------|------------|
-| coordinator | [roles/coordinator/role.md](roles/coordinator/role.md) | — | — |
-| strategist | [roles/strategist/role.md](roles/strategist/role.md) | STRATEGY-* | false |
-| generator | [roles/generator/role.md](roles/generator/role.md) | TESTGEN-* | true |
-| executor | [roles/executor/role.md](roles/executor/role.md) | TESTRUN-* | true |
-| analyst | [roles/analyst/role.md](roles/analyst/role.md) | TESTANA-* | false |
-
-## Role Router
-
-Parse `$ARGUMENTS`:
-- Has `--role <name>` -> Read `roles/<name>/role.md`, execute Phase 2-4
-- No `--role` -> `roles/coordinator/role.md`, execute entry router
-
-## Delegation Lock
-
-**Coordinator is a PURE ORCHESTRATOR. It coordinates, it does NOT do.**
-
-Before calling ANY tool, apply this check:
-
-| Tool Call | Verdict | Reason |
-|-----------|---------|--------|
-| `spawn_agent`, `wait_agent`, `close_agent`, `send_message`, `followup_task` | ALLOWED | Orchestration |
-| `list_agents` | ALLOWED | Agent health check |
-| `request_user_input` | ALLOWED | User interaction |
-| `mcp__maestro-tools__team_msg` | ALLOWED | Message bus |
-| `Read/Write` on `.workflow/.team/` files | ALLOWED | Session state |
-| `Read` on `roles/`, `commands/`, `specs/` | ALLOWED | Loading own instructions |
-| `Read/Grep/Glob` on project source code | BLOCKED | Delegate to worker |
-| `Edit` on any file outside `.workflow/` | BLOCKED | Delegate to worker |
-| `Bash("maestro delegate ...")` | BLOCKED | Only workers call CLI |
-| `Bash` running build/test/lint commands | BLOCKED | Delegate to worker |
-
-**If a tool call is BLOCKED**: STOP. Create a task, spawn a worker.
-
-**No exceptions for "simple" tasks.** Even a single-file read-and-report MUST go through spawn_agent.
-
----
-
-## Shared Constants
-
-- **Session prefix**: `TST`
-- **Session path**: `.workflow/.team/TST-<slug>-<date>/`
-- **Team name**: `testing`
-- **CLI tools**: `maestro delegate --mode analysis` (read-only), `maestro delegate --mode write` (modifications)
-- **Message bus**: `mcp__maestro-tools__team_msg(session_id=<session-id>, ...)`
-
-## Worker Spawn Template
-
-Coordinator spawns workers using this template:
-
-```
-spawn_agent({
-  agent_type: "team_worker",
-  task_name: "<task-id>",
-  fork_turns: "none",
-  message: `## Role Assignment
-role: <role>
-role_spec: <skill_root>/roles/<role>/role.md
-session: <session-folder>
-session_id: <session-id>
-requirement: <task-description>
-inner_loop: <true|false>
-
-Read role_spec file (<skill_root>/roles/<role>/role.md) to load Phase 2-4 domain instructions.
-
-## Task Context
-task_id: <task-id>
-title: <task-title>
-description: <task-description>
-pipeline_phase: <pipeline-phase>
-
-## Upstream Context
-<prev_context>`
-})
-```
-
-After spawning, use `wait_agent({ timeout_ms: 1800000 })` to collect results (30 min). If `result.timed_out`, send STATUS_CHECK via followup_task (wait 3 min), then FINALIZE with interrupt (wait 3 min), then mark timed_out and close agents. Use `close_agent({ target })` each worker.
-
-
-### Model Selection Guide
-
-| Role | model | reasoning_effort | Rationale |
-|------|-------|-------------------|-----------|
-| Strategist (STRATEGY-*) | (default) | high | Test strategy requires deep code understanding |
-| Generator (TESTGEN-*) | (default) | high | Test code generation needs precision |
-| Executor (TESTRUN-*) | (default) | medium | Running tests and collecting results, less reasoning |
-| Analyst (TESTANA-*) | (default) | high | Coverage analysis and quality assessment |
-
-Override model/reasoning_effort in spawn_agent when cost optimization is needed:
-```
-spawn_agent({
-  agent_type: "team_worker",
-  task_name: "<task-id>",
-  fork_turns: "none",
-  model: "<model-override>",
-  reasoning_effort: "<effort-level>",
-  message: "..."
-})
-```
-
-## User Commands
-
-| Command | Action |
-|---------|--------|
-| `check` / `status` | View pipeline status graph |
-| `resume` / `continue` | Advance to next step |
-| `revise <TASK-ID>` | Revise specific task |
-| `feedback <text>` | Inject feedback for revision |
-
-## v4 Agent Coordination
-
-### Message Semantics
-
-| Intent | API | Example |
-|--------|-----|---------|
-| Send strategy to running generators | `send_message` | Queue test strategy findings to TESTGEN-* workers |
-| Not used in this skill | `followup_task` | No resident agents -- all workers are one-shot |
-| Check running agents | `list_agents` | Verify parallel generator/executor health |
-
-### Parallel Test Generation
-
-Comprehensive pipeline spawns multiple generators (per layer) and executors in parallel:
-
-```
-// Spawn parallel generators for L1 and L2
-const genNames = ["TESTGEN-001", "TESTGEN-002"]
-for (const name of genNames) {
-  spawn_agent({ agent_type: "team_worker", task_name: name, ... })
-}
-wait_agent({ timeout_ms: 1800000 })  // 30 min
-```
-
-### GC Loop Coordination
-
-Generator-Critic loops create dynamic TESTGEN-fix and TESTRUN-fix tasks. The coordinator tracks `gc_rounds[layer]` and creates fix tasks dynamically when coverage is below target.
-
-### Agent Health Check
-
-Use `list_agents({})` in handleResume and handleComplete:
-
-```
-// Reconcile session state with actual running agents
-const running = list_agents({})
-// Compare with tasks.json active_agents
-// Reset orphaned tasks (in_progress but agent gone) to pending
-```
-
-### Named Agent Targeting
-
-Workers are spawned with `task_name: "<task-id>"` enabling direct addressing:
-- `send_message({ target: "TESTGEN-001", message: "..." })` -- queue strategy context to running generator
-- `close_agent({ target: "TESTRUN-001" })` -- cleanup by name after wait_agent returns
-
-## Completion Action
-
-When pipeline completes, coordinator presents:
-
-```
-request_user_input({
-  questions: [{
-    question: "Testing pipeline complete. What would you like to do?",
-    header: "Completion",
-    multiSelect: false,
-    options: [
-      { label: "Archive & Clean (Recommended)", description: "Archive session, clean up team" },
-      { label: "Keep Active", description: "Keep session for follow-up work" },
-      { label: "Deepen Coverage", description: "Add more test layers or increase coverage targets" }
-    ]
-  }]
-})
-```
-
-## Session Directory
-
-```
-.workflow/.team/TST-<slug>-<date>/
-├── .msg/messages.jsonl     # Team message bus
-├── .msg/meta.json          # Session metadata
-├── wisdom/                 # Cross-task knowledge
-├── strategy/               # Strategist output
-├── tests/                  # Generator output (L1-unit/, L2-integration/, L3-e2e/)
-├── results/                # Executor output
-└── analysis/               # Analyst output
-```
-
-## Specs Reference
-
-- [specs/pipelines.md](specs/pipelines.md) — Pipeline definitions and task registry
-- [specs/team-config.json](specs/team-config.json) — Team configuration
-
-## Error Handling
-
-| Scenario | Resolution |
-|----------|------------|
-| Unknown --role value | Error with available role list |
-| Role not found | Error with expected path (roles/<name>/role.md) |
-| CLI tool fails | Worker fallback to direct implementation |
-| GC loop exceeded | Accept current coverage with warning |
-| Fast-advance conflict | Coordinator reconciles on next callback |
-| Completion action fails | Default to Keep Active |
+---
+name: team-testing
+description: Unified team skill for testing team. Progressive test coverage through Generator-Critic loops, shared memory, and dynamic layer selection. Triggers on "team testing".
+allowed-tools: spawn_agent(*), wait_agent(*), send_message(*), followup_task(*), close_agent(*), list_agents(*), report_agent_job_result(*), request_user_input(*), Read(*), Write(*), Edit(*), Bash(*), Glob(*), Grep(*), mcp__maestro-tools__team_msg(*)
+---
+
+<purpose>
+Orchestrate multi-agent test pipeline: strategist -> generator -> executor -> analyst. Progressive layer coverage (L1/L2/L3) with Generator-Critic loops for coverage convergence.
+
+```
+Skill(skill="team-testing", args="task description")
+                    |
+         SKILL.md (this file) = Router
+                    |
+     +--------------+--------------+
+     |                             |
+  no --role flag              --role <name>
+     |                             |
+  Coordinator                  Worker
+  roles/coordinator/role.md    roles/<name>/role.md
+     |
+     +-- analyze -> dispatch -> spawn workers -> STOP
+                                    |
+                    +-------+-------+-------+-------+
+                    v       v       v       v
+                [strat] [gen]  [exec]  [analyst]
+                team-worker agents, each loads roles/<role>/role.md
+```
+</purpose>
+
+<context>
+### Role Registry
+
+| Role | Path | Prefix | Inner Loop |
+|------|------|--------|------------|
+| coordinator | [roles/coordinator/role.md](roles/coordinator/role.md) | -- | -- |
+| strategist | [roles/strategist/role.md](roles/strategist/role.md) | STRATEGY-* | false |
+| generator | [roles/generator/role.md](roles/generator/role.md) | TESTGEN-* | true |
+| executor | [roles/executor/role.md](roles/executor/role.md) | TESTRUN-* | true |
+| analyst | [roles/analyst/role.md](roles/analyst/role.md) | TESTANA-* | false |
+
+### Role Router
+
+Parse `$ARGUMENTS`:
+- Has `--role <name>` -> Read `roles/<name>/role.md`, execute Phase 2-4
+- No `--role` -> `roles/coordinator/role.md`, execute entry router
+
+### Delegation Lock
+
+**Coordinator is a PURE ORCHESTRATOR. It coordinates, it does NOT do.**
+
+Before calling ANY tool, apply this check:
+
+| Tool Call | Verdict | Reason |
+|-----------|---------|--------|
+| `spawn_agent`, `wait_agent`, `close_agent`, `send_message`, `followup_task` | ALLOWED | Orchestration |
+| `list_agents` | ALLOWED | Agent health check |
+| `request_user_input` | ALLOWED | User interaction |
+| `mcp__maestro-tools__team_msg` | ALLOWED | Message bus |
+| `Read/Write` on `.workflow/.team/` files | ALLOWED | Session state |
+| `Read` on `roles/`, `commands/`, `specs/` | ALLOWED | Loading own instructions |
+| `Read/Grep/Glob` on project source code | BLOCKED | Delegate to worker |
+| `Edit` on any file outside `.workflow/` | BLOCKED | Delegate to worker |
+| `Bash("maestro delegate ...")` | BLOCKED | Only workers call CLI |
+| `Bash` running build/test/lint commands | BLOCKED | Delegate to worker |
+
+**If a tool call is BLOCKED**: STOP. Create a task, spawn a worker.
+
+**No exceptions for "simple" tasks.** Even a single-file read-and-report MUST go through spawn_agent.
+
+### Shared Constants
+
+- **Session prefix**: `TST`
+- **Session path**: `.workflow/.team/TST-<slug>-<date>/`
+- **Team name**: `testing`
+- **CLI tools**: `maestro delegate --mode analysis` (read-only), `maestro delegate --mode write` (modifications)
+- **Message bus**: `mcp__maestro-tools__team_msg(session_id=<session-id>, ...)`
+
+### Worker Spawn Template
+
+Coordinator spawns workers using this template:
+
+```
+spawn_agent({
+  agent_type: "team_worker",
+  task_name: "<task-id>",
+  fork_turns: "none",
+  message: `## Role Assignment
+role: <role>
+role_spec: <skill_root>/roles/<role>/role.md
+session: <session-folder>
+session_id: <session-id>
+requirement: <task-description>
+inner_loop: <true|false>
+
+Read role_spec file (<skill_root>/roles/<role>/role.md) to load Phase 2-4 domain instructions.
+
+## Task Context
+task_id: <task-id>
+title: <task-title>
+description: <task-description>
+pipeline_phase: <pipeline-phase>
+
+## Upstream Context
+<prev_context>`
+})
+```
+
+After spawning, use `wait_agent({ timeout_ms: 1800000 })` to collect results (30 min). If `result.timed_out`, send STATUS_CHECK via followup_task (wait 3 min), then FINALIZE with interrupt (wait 3 min), then mark timed_out and close agents. Use `close_agent({ target })` each worker.
+
+### Model Selection Guide
+
+| Role | model | reasoning_effort | Rationale |
+|------|-------|-------------------|-----------|
+| Strategist (STRATEGY-*) | (default) | high | Test strategy requires deep code understanding |
+| Generator (TESTGEN-*) | (default) | high | Test code generation needs precision |
+| Executor (TESTRUN-*) | (default) | medium | Running tests and collecting results, less reasoning |
+| Analyst (TESTANA-*) | (default) | high | Coverage analysis and quality assessment |
+
+Override model/reasoning_effort in spawn_agent when cost optimization is needed:
+```
+spawn_agent({
+  agent_type: "team_worker",
+  task_name: "<task-id>",
+  fork_turns: "none",
+  model: "<model-override>",
+  reasoning_effort: "<effort-level>",
+  message: "..."
+})
+```
+
+### User Commands
+
+| Command | Action |
+|---------|--------|
+| `check` / `status` | View pipeline status graph |
+| `resume` / `continue` | Advance to next step |
+| `revise <TASK-ID>` | Revise specific task |
+| `feedback <text>` | Inject feedback for revision |
+
+### v4 Agent Coordination
+
+#### Message Semantics
+
+| Intent | API | Example |
+|--------|-----|---------|
+| Send strategy to running generators | `send_message` | Queue test strategy findings to TESTGEN-* workers |
+| Not used in this skill | `followup_task` | No resident agents -- all workers are one-shot |
+| Check running agents | `list_agents` | Verify parallel generator/executor health |
+
+#### Parallel Test Generation
+
+Comprehensive pipeline spawns multiple generators (per layer) and executors in parallel:
+
+```
+// Spawn parallel generators for L1 and L2
+const genNames = ["TESTGEN-001", "TESTGEN-002"]
+for (const name of genNames) {
+  spawn_agent({ agent_type: "team_worker", task_name: name, ... })
+}
+wait_agent({ timeout_ms: 1800000 })  // 30 min
+```
+
+#### GC Loop Coordination
+
+Generator-Critic loops create dynamic TESTGEN-fix and TESTRUN-fix tasks. The coordinator tracks `gc_rounds[layer]` and creates fix tasks dynamically when coverage is below target.
+
+#### Agent Health Check
+
+Use `list_agents({})` in handleResume and handleComplete:
+
+```
+// Reconcile session state with actual running agents
+const running = list_agents({})
+// Compare with tasks.json active_agents
+// Reset orphaned tasks (in_progress but agent gone) to pending
+```
+
+#### Named Agent Targeting
+
+Workers are spawned with `task_name: "<task-id>"` enabling direct addressing:
+- `send_message({ target: "TESTGEN-001", message: "..." })` -- queue strategy context to running generator
+- `close_agent({ target: "TESTRUN-001" })` -- cleanup by name after wait_agent returns
+
+### Completion Action
+
+When pipeline completes, coordinator presents:
+
+```
+request_user_input({
+  questions: [{
+    question: "Testing pipeline complete. What would you like to do?",
+    header: "Completion",
+    multiSelect: false,
+    options: [
+      { label: "Archive & Clean (Recommended)", description: "Archive session, clean up team" },
+      { label: "Keep Active", description: "Keep session for follow-up work" },
+      { label: "Deepen Coverage", description: "Add more test layers or increase coverage targets" }
+    ]
+  }]
+})
+```
+
+### Session Directory
+
+```
+.workflow/.team/TST-<slug>-<date>/
++-- .msg/messages.jsonl     # Team message bus
++-- .msg/meta.json          # Session metadata
++-- wisdom/                 # Cross-task knowledge
++-- strategy/               # Strategist output
++-- tests/                  # Generator output (L1-unit/, L2-integration/, L3-e2e/)
++-- results/                # Executor output
++-- analysis/               # Analyst output
+```
+
+### Specs Reference
+
+- [specs/pipelines.md](specs/pipelines.md) -- Pipeline definitions and task registry
+- [specs/team-config.json](specs/team-config.json) -- Team configuration
+</context>
+
+<error_codes>
+
+| Scenario | Resolution |
+|----------|------------|
+| Unknown --role value | Error with available role list |
+| Role not found | Error with expected path (roles/<name>/role.md) |
+| CLI tool fails | Worker fallback to direct implementation |
+| GC loop exceeded | Accept current coverage with warning |
+| Fast-advance conflict | Coordinator reconciles on next callback |
+| Completion action fails | Default to Keep Active |
+</error_codes>
+
+<success_criteria>
+- [ ] Role router correctly dispatches to coordinator or worker based on --role flag
+- [ ] Pipeline executes strategist -> generator -> executor -> analyst in order
+- [ ] Parallel generators spawned per test layer (L1/L2/L3)
+- [ ] GC loops create fix tasks dynamically when coverage is below target
+- [ ] Session state persisted after each wave
+- [ ] Completion action presented and handled correctly
+</success_criteria>

package/.codex/skills/wiki-connect/SKILL.md

@@ -0,0 +1,75 @@
+---
+name: wiki-connect
+description: Wiki knowledge graph link discovery and health improvement. Finds orphaned entries, missing connections, transitive gaps. Scores candidates and optionally auto-applies new related links via --fix.
+argument-hint: "[--scope <type>] [--min-similarity N] [--fix] [--max N]"
+allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+---
+
+<purpose>
+Knowledge graph link discovery. Analyzes wiki index to find orphaned entries, missing
+bidirectional links, and transitive closure gaps. Scores connection candidates and
+optionally auto-applies new `related` links to improve graph connectivity.
+</purpose>
+
+<context>
+$ARGUMENTS — optional flags.
+
+**Flags:**
+- `--scope <type>` — Limit to wiki type (spec, phase, memory, note, lesson, issue). Default: all.
+- `--min-similarity N` — Threshold 0.0-1.0 (default: 0.3)
+- `--fix` — Auto-apply top suggestions
+- `--max N` — Max suggestions (default: 20)
+
+**Output**: `.workflow/learning/wiki-connections-{date}.md`
+</context>
+
+<execution>
+
+### Stage 1: Load Wiki State
+Parallel `maestro wiki` commands: `list --json`, `health`, `orphans`, `hubs --top 10`.
+
+### Stage 2: Identify Connection Candidates
+- **Orphan rescue**: BM25 search by title, tag overlap, same phase
+- **Missing bidirectional**: A→B exists but B→A missing
+- **Transitive closure**: A→B and B→C but no A→C (with shared tags/phase)
+- **Type bridge**: Different types referencing same concept but unlinked
+
+### Stage 3: Score Candidates
+```
+score = 0.4 × tag_overlap + 0.3 × title_bm25 + 0.2 × same_phase + 0.1 × type_bridge
+```
+Filter by `--min-similarity`, rank desc, limit by `--max`.
+
+### Stage 4: Present Suggestions
+Display ranked suggestions with scores, reasons, projected health delta.
+If not `--fix`: display and exit.
+
+### Stage 5: Apply (--fix only)
+For each suggestion: get entry → append target to `related` → update via `maestro wiki update`.
+Re-run `maestro wiki health` for delta.
+
+### Stage 6: Persist
+Write `wiki-connections-{date}.md`. Append graph insights to `lessons.jsonl` (source: "wiki-connect").
+
+**Next steps:** `/wiki-digest <topic>`, `/learn-follow <wiki-id>`, `maestro wiki graph`
+</execution>
+
+<error_codes>
+| Code | Severity | Condition | Recovery |
+|------|----------|-----------|----------|
+| E001 | error | No wiki entries found | Initialize wiki content |
+| W001 | warning | No candidates above threshold | Lower --min-similarity |
+| W002 | warning | Some wiki updates failed during --fix | Retry manually |
+| W003 | warning | Health score unchanged after fix | Connections may not affect specific metrics |
+</error_codes>
+
+<success_criteria>
+- [ ] Wiki index loaded with type distribution
+- [ ] Baseline health score recorded
+- [ ] Orphans identified and rescue candidates generated
+- [ ] Candidates scored and ranked
+- [ ] Suggestions displayed with scores and reasons
+- [ ] If --fix: entries updated, new health score reported
+- [ ] Report written to `wiki-connections-{date}.md`
+- [ ] Graph insights appended to `lessons.jsonl`
+</success_criteria>

package/.codex/skills/wiki-digest/SKILL.md

@@ -0,0 +1,87 @@
+---
+name: wiki-digest
+description: Knowledge synthesis from wiki entries. Theme clustering, gap analysis, coverage heatmap (type × theme matrix). Optionally creates knowledge-gap issues. Persists meta-insights to lessons.jsonl.
+argument-hint: "[<topic>|--recent N] [--type <type>] [--format brief|full] [--create-issues]"
+allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+---
+
+<purpose>
+Knowledge synthesis that generates actionable digests from the wiki knowledge graph.
+Clusters entries by semantic theme, identifies knowledge gaps, and produces a coverage
+heatmap. Unlike `maestro wiki list` (raw entries), this synthesizes and interprets
+the knowledge base with gap analysis and recommended actions.
+</purpose>
+
+<context>
+$ARGUMENTS — scope and optional flags.
+
+**Scope resolution:**
+- `<topic>` — Search wiki for matching entries
+- `--recent N` — Entries updated in last N days
+- `--type <type>` — Filter by wiki type
+- No args — entire wiki
+
+**Flags:**
+- `--format brief` — Compact summary (default)
+- `--format full` — Detailed with per-entry summaries
+- `--create-issues` — Auto-create knowledge-gap issues in issues.jsonl
+
+**Output**: `.workflow/learning/digest-{slug}-{date}.md`
+</context>
+
+<execution>
+
+### Stage 1: Scope & Load
+Load entries via `maestro wiki list/search`. Run `maestro wiki health` for baseline.
+
+### Stage 2: Theme Clustering
+Group entries into 3-5 themes via: tag co-occurrence, title BM25 similarity, relationship proximity, type grouping.
+
+### Stage 3: Per-Theme Analysis
+Per theme: summary paragraph, key entries (by hub score), gap detection (broken links, orphans, TODO markers, missing perspectives), health score.
+
+### Stage 4: Cross-Reference with Lessons
+Search `lessons.jsonl` for related insights. Flag unlinked insights (lessons matching theme but not referenced by wiki entries).
+
+### Stage 5: Coverage Heatmap
+Type × theme matrix showing knowledge density:
+```
+              Theme 1    Theme 2    Theme 3
+spec          ███░░      ░░░░░      █████
+phase         ████░      ███░░      ░░░░░
+lesson        █░░░░      ██░░░      ████░
+```
+Empty cells = knowledge gaps.
+
+### Stage 6: Write Digest
+Produce `digest-{slug}-{date}.md` with themes, heatmap, gaps, unlinked insights, recommended actions.
+
+### Stage 7: Gap → Issue (if --create-issues)
+For each gap: dedup against issues.jsonl, append with `type: "knowledge-gap"`, `source: "wiki-digest"`.
+
+### Stage 8: Persist
+Append meta-insights to `lessons.jsonl` (source: "wiki-digest"). Display summary.
+
+**Next steps:** `/learn-follow <wiki-id>`, `/wiki-connect --fix`, `/learn-decompose <path>`
+</execution>
+
+<error_codes>
+| Code | Severity | Condition | Recovery |
+|------|----------|-----------|----------|
+| E001 | error | No wiki entries found | Initialize wiki content |
+| E002 | error | Topic search returned 0 | Broaden topic |
+| W001 | warning | Too few entries (<5) | Themes may be trivial |
+| W002 | warning | lessons.jsonl not found | Skip cross-reference |
+| W003 | warning | Some entry bodies failed to load | Partial summaries |
+</error_codes>
+
+<success_criteria>
+- [ ] Scope parsed and entries loaded
+- [ ] Entries clustered into 3-5 semantic themes
+- [ ] Per-theme analysis with gaps identified
+- [ ] Cross-reference with lessons.jsonl completed
+- [ ] Coverage heatmap generated
+- [ ] If --create-issues: gap issues created (deduped)
+- [ ] Digest written to `digest-{slug}-{date}.md`
+- [ ] Meta-insights appended to lessons.jsonl
+</success_criteria>