maestro-flow 0.3.14 → 0.3.16

package/.codex/skills/manage-codebase-rebuild/SKILL.md

@@ -157,25 +157,15 @@ Single wave generates `wave-1.csv`. No `prev_context` needed (all tasks independ
 
 ### Session Initialization
 
-```javascript
-const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
-
-// Parse flags
-const AUTO_YES = $ARGUMENTS.includes('--yes') || $ARGUMENTS.includes('-y')
-const continueMode = $ARGUMENTS.includes('--continue')
-const concurrencyMatch = $ARGUMENTS.match(/(?:--concurrency|-c)\s+(\d+)/)
-const maxConcurrency = concurrencyMatch ? parseInt(concurrencyMatch[1]) : 5
-
-// Parse rebuild-specific flags
-const forceMode = $ARGUMENTS.includes('--force') || AUTO_YES
-const skipCommit = $ARGUMENTS.includes('--skip-commit')
+Parse `$ARGUMENTS` to extract:
+- `AUTO_YES` from `--yes` / `-y`
+- `continueMode` from `--continue`
+- `maxConcurrency` from `--concurrency N` / `-c N` (default: 5)
+- `forceMode` from `--force` (or implied by AUTO_YES)
+- `skipCommit` from `--skip-commit`
 
-const dateStr = getUtc8ISOString().substring(0, 10).replace(/-/g, '')
-const sessionId = `${dateStr}-rebuild-full`
-const sessionFolder = `.workflow/.csv-wave/${sessionId}`
-
-Bash(`mkdir -p ${sessionFolder}`)
-```
+Session ID: `{YYYYMMDD}-rebuild-full`
+Session folder: `.workflow/.csv-wave/{sessionId}/` — create via `mkdir -p`
 
 ### Phase 1: Setup -> CSV
 
@@ -183,36 +173,18 @@ Bash(`mkdir -p ${sessionFolder}`)
 
 **Steps**:
 
-1. **Validate .workflow/ exists**:
-   - Check `.workflow/state.json` exists
-   - If not: abort with "Run init first"
-
-2. **Confirm rebuild**:
-   - If `.workflow/codebase/` exists AND NOT forceMode:
-     ```
-     AskUserQuestion: "Codebase docs already exist. Rebuild will overwrite all files. Continue? [y/N]"
-     If no: exit
-     ```
-   - If forceMode or confirmed: clear `.workflow/codebase/`
-
-3. **Prepare directory structure**:
+1. **Validate** `.workflow/state.json` exists — abort with "Run init first" if missing
+2. **Confirm rebuild**: If `.workflow/codebase/` exists AND NOT forceMode, prompt user. If forceMode or confirmed, clear `.workflow/codebase/`
+3. **Prepare directories**:
    ```bash
    mkdir -p .workflow/codebase/tech-registry
    mkdir -p .workflow/codebase/feature-maps
    mkdir -p .workflow/codebase/action-logs
    ```
-
-4. **Detect source directories**:
-   - Check for: `src/`, `lib/`, `app/`, `packages/`
-   - Read `project.md` Tech Stack section if available for context
-   - If no source directories found: abort with "No source files in project"
-
-5. **Load project specs** (if available):
-   - Read `.workflow/specs/` for architecture context
-
-6. **Generate tasks.csv**: 5 rows, all wave 1, no dependencies.
-
-7. **User validation**: Display doc generator breakdown. Skip if AUTO_YES.
+4. **Detect source directories**: `src/`, `lib/`, `app/`, `packages/` — abort if none found
+5. **Load project specs** from `.workflow/specs/` if available
+6. **Generate tasks.csv**: 5 rows, all wave 1, no dependencies
+7. **User validation**: Display doc generator breakdown (skip if AUTO_YES)
 
 ### Phase 2: Wave Execution (Single Wave)
 
@@ -220,11 +192,7 @@ Bash(`mkdir -p ${sessionFolder}`)
 
 #### Wave 1: All Generators (Parallel)
 
-1. Read master `tasks.csv`
-2. Filter rows where `wave == 1` AND `status == pending`
-3. No prev_context needed (single wave, all independent)
-4. Write `wave-1.csv`
-5. Execute:
+Filter master `tasks.csv` for `wave == 1 AND status == pending` → write `wave-1.csv` (no prev_context needed, all independent).
 
 ```javascript
 spawn_agents_on_csv({
@@ -247,21 +215,15 @@ spawn_agents_on_csv({
 })
 ```
 
-6. Read `wave-1-results.csv`, merge into master `tasks.csv`
-7. Delete `wave-1.csv`
+Merge `wave-1-results.csv` into master `tasks.csv`, delete `wave-1.csv`.
 
 ### Phase 3: Results -> .workflow/codebase/
 
 **Objective**: Assemble doc-index.json from agent findings, validate, update state.
 
-1. Read final master `tasks.csv`
-2. Export as `results.csv`
+Export master `tasks.csv` as `results.csv`.
 
-3. **Assemble doc-index.json**:
-   - Read component findings from task 1 (Component Scanner)
-   - Read feature findings from task 2 (Feature Mapper)
-   - Read requirement/ADR findings from task 3 (Requirement Linker)
-   - Merge into complete doc-index.json:
+**Assemble doc-index.json** by merging findings from tasks 1-3 (Component Scanner, Feature Mapper, Requirement Linker):
    ```json
    {
      "version": "1.0",
@@ -277,15 +239,11 @@ spawn_agents_on_csv({
    ```
    - Write to `.workflow/codebase/doc-index.json`
 
-4. **Validate output files**:
-   - Check doc-index.json exists and is valid JSON
-   - Check tech-registry/_index.md exists (from task 4)
-   - Check feature-maps/_index.md exists (from task 5)
-   - Log warnings for any missing files
+**Validate output files**: doc-index.json (valid JSON), tech-registry/_index.md, feature-maps/_index.md — log warnings for missing.
 
-5. **Update state.json**: Set `codebase.last_rebuild` timestamp.
+**Update state.json**: Set `codebase.last_rebuild` timestamp.
 
-6. **Generate context.md**:
+**Generate context.md**:
 
 ```markdown
 # Codebase Rebuild Report
@@ -315,11 +273,9 @@ spawn_agents_on_csv({
 - Run manage-codebase-refresh for future incremental updates
 ```
 
-7. **Auto-commit** (unless --skip-commit):
-   - Stage `.workflow/codebase/` files
-   - Suggest commit: "docs(codebase): full rebuild of codebase documentation"
+**Auto-commit** (unless --skip-commit): Stage `.workflow/codebase/` files, commit "docs(codebase): full rebuild of codebase documentation".
 
-8. **Display completion report**:
+**Display completion report**:
 
 ```
 === CODEBASE REBUILD COMPLETE ===
@@ -357,17 +313,11 @@ Next steps:
 
 #### Protocol
 
-1. **Read** `{session_folder}/discoveries.ndjson` before own analysis
-2. **Skip covered**: If discovery of same type + dedup key exists, skip
-3. **Write immediately**: Append findings as discovered
-4. **Append-only**: Never modify or delete
-5. **Deduplicate**: Check before writing
+Read `{session_folder}/discoveries.ndjson` before own analysis. Deduplicate by type + dedup key before writing. Append-only — never modify or delete. Generators share discoveries to skip redundant scanning.
 
 ```bash
 echo '{"ts":"<ISO>","worker":"1","type":"tech_stack","data":{"framework":"Express","language":"TypeScript","tools":["jest","eslint","prettier"]}}' >> {session_folder}/discoveries.ndjson
 ```
-
-Generators share discoveries so other generators can skip redundant scanning (e.g., Component Scanner discovers components, Feature Mapper and Tech Registry Writer can leverage those findings).
 </execution>
 
 <error_codes>

package/.codex/skills/manage-codebase-refresh/SKILL.md

@@ -28,50 +28,23 @@ $manage-codebase-refresh "--since 3d --deep"
 
 ### Step 1: Validate Preconditions
 
-```bash
-test -d .workflow || exit 1        # E001: not initialized
-test -d .workflow/codebase || exit 1  # E002: no docs, use codebase-rebuild
-```
+Verify `.workflow/` exists (E001) and `.workflow/codebase/` exists (E002 -- use codebase-rebuild instead).
 
 ### Step 2: Detect Changes
 
-Determine baseline timestamp:
-1. If `--since` provided, use that date
-2. Else read `state.json` field `codebase_last_refreshed` or `codebase_last_rebuilt`
-3. Fallback to last 7 days
-
-```bash
-git diff --name-only --since="{baseline}" HEAD
-```
-
-If no changes detected, report clean state (W001) and exit.
+Resolve baseline: `--since` flag > `state.json.codebase_last_refreshed` > `codebase_last_rebuilt` > 7-day fallback. Run `git diff --name-only --since="{baseline}" HEAD`. If no changes: W001, exit.
 
 ### Step 3: Map Changes to Docs
 
-Read `.workflow/codebase/doc-index.json` to find which documentation entries cover the changed files.
-Build a list of affected doc entries that need refresh.
+Read `.workflow/codebase/doc-index.json` to find doc entries covering changed files. Build affected entry list.
 
 ### Step 4: Refresh Affected Docs
 
-For each affected documentation entry:
-1. Re-read the source files that changed
-2. Update the corresponding doc section in `.workflow/codebase/`
-3. Update the entry's timestamp in `doc-index.json`
-
-If `--deep` flag is set, also re-scan adjacent files for context changes.
+For each affected entry: re-read changed source files, update corresponding doc in `.workflow/codebase/`, update timestamp in `doc-index.json`. With `--deep`: also re-scan adjacent files.
 
 ### Step 5: Update State
 
-Update `doc-index.json` timestamps for all refreshed entries.
-Update `state.json` with `codebase_last_refreshed: "{ISO timestamp}"`.
-
-Display summary:
-```
-=== CODEBASE REFRESH ===
-Changes detected: {N} files
-Docs refreshed: {M} entries
-Skipped (unchanged): {K} entries
-```
+Update `doc-index.json` timestamps and `state.json.codebase_last_refreshed`. Display summary with change/refresh/skip counts.
 </execution>
 
 <error_codes>

package/.codex/skills/manage-issue/SKILL.md

@@ -34,12 +34,7 @@ If missing or invalid, display usage and prompt user (E_NO_SUBCOMMAND, E_INVALID
 
 ### Step 2: Ensure Storage
 
-```bash
-mkdir -p .workflow/issues
-touch .workflow/issues/issues.jsonl 2>/dev/null
-```
-
-Auto-create directory and empty file if missing (E_ISSUES_DIR_MISSING handled silently).
+Auto-create `.workflow/issues/` and empty `issues.jsonl` if missing (E_ISSUES_DIR_MISSING handled silently).
 
 ### Step 3: Execute Subcommand
 

package/.codex/skills/manage-issue-discover/SKILL.md

@@ -163,34 +163,21 @@ Also writes to:
 ### Session Initialization
 
 ```javascript
-const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
-
-// Parse flags
-const AUTO_YES = $ARGUMENTS.includes('--yes') || $ARGUMENTS.includes('-y')
-const continueMode = $ARGUMENTS.includes('--continue')
-const concurrencyMatch = $ARGUMENTS.match(/(?:--concurrency|-c)\s+(\d+)/)
-const maxConcurrency = concurrencyMatch ? parseInt(concurrencyMatch[1]) : 8
-
-// Parse mode
-const cleanArgs = $ARGUMENTS
-  .replace(/--yes|-y|--continue|--concurrency\s+\d+|-c\s+\d+/g, '')
-  .trim()
-
-const isByPrompt = cleanArgs.startsWith('by-prompt')
-const userPrompt = isByPrompt
-  ? cleanArgs.replace(/^by-prompt\s*/, '').replace(/^['"]|['"]$/g, '').trim()
-  : ''
-const mode = isByPrompt ? 'by-prompt' : 'multi'
-
-const dateStr = getUtc8ISOString().substring(0, 10).replace(/-/g, '')
-const timeStr = getUtc8ISOString().substring(11, 19).replace(/:/g, '')
-const sessionId = `DBP-${dateStr}-${timeStr}`
-const csvSessionId = `${dateStr}-discover-${mode}`
-const sessionFolder = `.workflow/.csv-wave/${csvSessionId}`
-const discoveryDir = `.workflow/issues/discoveries/${sessionId}`
-
-Bash(`mkdir -p ${sessionFolder} && mkdir -p ${discoveryDir} && mkdir -p .workflow/issues`)
-Bash(`touch .workflow/issues/issues.jsonl`)
+// Parse from $ARGUMENTS:
+//   AUTO_YES   <- --yes | -y
+//   continueMode <- --continue
+//   maxConcurrency <- --concurrency N | -c N  (default: 8)
+//   mode       <- "by-prompt" if clean args start with "by-prompt", else "multi"
+//   userPrompt <- text after "by-prompt" (stripped of quotes)
+
+// Session IDs (UTC+8):
+//   sessionId    = DBP-{YYYYMMDD}-{HHmmss}
+//   csvSessionId = {YYYYMMDD}-discover-{mode}
+//   sessionFolder = .workflow/.csv-wave/{csvSessionId}
+//   discoveryDir  = .workflow/issues/discoveries/{sessionId}
+
+// Create: sessionFolder, discoveryDir, .workflow/issues/
+// Touch: .workflow/issues/issues.jsonl
 ```
 
 Initialize `discovery-state.json`:
@@ -262,24 +249,17 @@ spawn_agents_on_csv({
   max_concurrency: maxConcurrency,
   max_runtime_seconds: 900,
   output_csv_path: `${sessionFolder}/wave-1-results.csv`,
-  output_schema: {
-    type: "object",
-    properties: {
-      id: { type: "string" },
-      status: { type: "string", enum: ["completed", "failed"] },
-      findings: { type: "string" },
-      issues_found: { type: "string" },
-      severity_distribution: { type: "string" },
-      error: { type: "string" }
-    },
-    required: ["id", "status", "findings"]
+  output_schema: { // required: id, status, findings
+    id: "string", status: "completed|failed",
+    findings: "string", issues_found: "string",
+    severity_distribution: "string", error: "string"
   }
 })
 ```
 
-6. Read `wave-1-results.csv`, merge into master `tasks.csv`
-7. For each completed perspective, save raw findings to `{discoveryDir}/{perspective}-findings.json`
-8. Update `discovery-state.json`: `perspectives_completed += ["{perspective}"]`
+6. Merge `wave-1-results.csv` into master `tasks.csv`
+7. Save per-perspective findings to `{discoveryDir}/{perspective}-findings.json`
+8. Update `discovery-state.json` with completed perspectives
 9. Delete `wave-1.csv`
 
 **Perspective scan agent protocol**:
@@ -308,20 +288,12 @@ spawn_agents_on_csv({
 8. Delete `wave-2.csv`
 
 **Dedup agent protocol**:
-- Load all perspective findings from prev_context
-- Merge into single list
-- Deduplicate:
-  - Group findings by affected file path
-  - Within each file group, compare descriptions
-  - If two findings describe the same issue (>80% overlap or same file:line), keep higher severity
-- For each unique finding:
-  - Generate ISS-YYYYMMDD-NNN ID (read existing issues.jsonl to avoid collisions)
-  - Build issue record with full schema (id, title, status, priority, severity, source, description, fix_direction, context, tags, etc.)
-  - Severity-to-priority: critical->1, high->2, medium->3, low->4
-  - Set source = "discovery", tags = ["{perspective}"]
-- Append all issues to `.workflow/issues/issues.jsonl`
-- Append to `{discoveryDir}/discovery-issues.jsonl`
-- Report: total issues_found (pre-dedup), issues after dedup, severity_distribution
+- Merge all perspective findings from prev_context into single list
+- Deduplicate: group by file path, compare descriptions (>80% overlap or same file:line → keep higher severity)
+- For each unique finding: generate `ISS-YYYYMMDD-NNN` ID (collision-safe), build full issue record
+  - Severity-to-priority: critical→1, high→2, medium→3, low→4; source = "discovery", tags = ["{perspective}"]
+- Append to `.workflow/issues/issues.jsonl` and `{discoveryDir}/discovery-issues.jsonl`
+- Report: pre-dedup count, post-dedup count, severity_distribution
 
 ### Phase 3: Results Aggregation
 
@@ -384,48 +356,16 @@ spawn_agents_on_csv({
 {list of ISS-YYYYMMDD-NNN IDs with titles}
 ```
 
-6. **Display summary**:
-
-```
-====================================================
-  DISCOVERY COMPLETE: {sessionId}
-  Mode: {mode} ({perspective_count} perspectives)
-  Findings: {issues_found} raw, {issues_deduplicated} unique
-  Issues created: {issues_deduplicated}
-====================================================
-
-BREAKDOWN BY PERSPECTIVE:
-  Security:        {count}
-  Performance:     {count}
-  Reliability:     {count}
-  Maintainability: {count}
-  Scalability:     {count}
-  UX:              {count}
-  Accessibility:   {count}
-  Compliance:      {count}
-
-BREAKDOWN BY SEVERITY:
-  Critical: {count}
-  High:     {count}
-  Medium:   {count}
-  Low:      {count}
-
-Files:
-  {session_folder}/results.csv
-  {session_folder}/context.md
-  {discoveryDir}/discovery-state.json
-  {discoveryDir}/discovery-issues.jsonl
-  .workflow/issues/issues.jsonl (appended)
-```
+6. **Display summary**: Session ID, mode, perspective count, raw vs unique finding counts, per-perspective breakdown, severity breakdown, and output file paths.
 
 7. **Next step routing**:
 
 | Result | Suggestion |
 |--------|------------|
-| Critical issues found | `$manage-issue "list --severity critical"` -- Review critical issues |
-| Issues created | `$manage-issue "list"` -- View all issues |
-| Specific area needs deeper look | `$manage-issue-discover "by-prompt '...'"` -- Explore deeper |
-| Full scan complete | `$manage-issue "list --source discovery"` -- View discovered issues |
+| Critical issues found | `$manage-issue "list --severity critical"` |
+| Issues created | `$manage-issue "list"` |
+| Specific area needs deeper look | `$manage-issue-discover "by-prompt '...'"` |
+| Full scan complete | `$manage-issue "list --source discovery"` |
 
 ### Shared Discovery Board Protocol
 

package/.codex/skills/{manage-memory → manage-knowhow}/SKILL.md

@@ -1,95 +1,95 @@
----
-name: manage-memory
-description: Manage memory entries across workflow and system stores (list, search, view, edit, delete, prune)
-argument-hint: "[list|search|view|edit|delete|prune] [query|id|file] [--store workflow|system|all] [--tag tag] [--type compact|tip]"
-allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
----
-
-<purpose>
-Manage memory entries across workflow and system stores. Provides list, search, view, edit, delete, and prune operations over `.workflow/memory/` (workflow store) and `~/.claude/projects/{project}/memory/` (system store).
-</purpose>
-
-<context>
-$ARGUMENTS — subcommand followed by options. Defaults to `list` if no arguments.
-
-```bash
-$manage-memory
-$manage-memory "list --store workflow"
-$manage-memory "search authentication"
-$manage-memory "view MEM-20260318-001"
-$manage-memory "edit MEMORY.md"
-$manage-memory "delete TIP-20260318-001 --confirm"
-$manage-memory "prune --before 2026-01-01 --type tip --dry-run"
-```
-
-**Subcommands**: `list`, `search`, `view`, `edit`, `delete`, `prune`.
-
-**Flags**:
-- `--store workflow|system|all` — Target store (default: all)
-- `--tag <tag>` — Filter by tag
-- `--type compact|tip` — Filter by entry type
-- `--confirm` — Skip delete confirmation prompt
-- `--before <date>` / `--after <date>` — Date filters for prune
-- `--dry-run` — Preview prune without deleting
-</context>
-
-<execution>
-
-### Step 1: Resolve Store Paths
-
-- **Workflow store**: `.workflow/memory/` (entries: `MEM-*.md`, `TIP-*.md`, indexed in unified `.workflow/wiki-index.json`)
-- **System store**: `~/.claude/projects/{project}/memory/` (files: `MEMORY.md` + topic `.md` files)
-
-Derive system path from project root (replace path separators with `--`, prefix drive letter).
-
-### Step 2: Parse Subcommand
-
-Default to `list` if no arguments. Parse first token as subcommand.
-
-### Step 3: Execute Subcommand
-
-**list**: Show entries from both stores (or filtered by `--store`, `--tag`, `--type`).
-- Workflow: use `maestro wiki list --type memory --json` or read `.workflow/wiki-index.json`, display ID, type, date, tags, title
-- System: list `.md` files in system memory directory
-
-**search `<query>`**: Full-text grep across both stores. Rank by match count.
-
-**view `<id|file>`**: Auto-detect store from format (`MEM-*/TIP-*` = workflow, else system). Display full content.
-
-**edit `<file>`**: Edit a system memory file. Read current content, apply changes. Warn if MEMORY.md exceeds 200 lines (W003).
-
-**delete `<id|file>`**: Require confirmation (or `--confirm` flag). MEMORY.md cannot be deleted (E004). Remove entry file (WikiIndexer auto-updates `.workflow/wiki-index.json` on next access).
-
-**prune**: Requires at least one filter (`--tag`, `--type`, `--before`, `--after`). Workflow store only. `--dry-run` previews without deleting.
-
-### Step 4: Integrity Check
-
-After write operations, verify:
-- No orphaned files without index entries (W001)
-- No dangling index references to missing files (W001)
-- System MEMORY.md references valid topic files (W002)
-</execution>
-
-<error_codes>
-| Code | Severity | Description |
-|------|----------|-------------|
-| E001 | error | No memory stores found -- run `Skill({ skill: "memory-capture" })` or create MEMORY.md |
-| E002 | error | Entry ID or filename not found |
-| E003 | error | Prune requires at least one filter flag |
-| E004 | error | Cannot delete MEMORY.md -- use `edit` subcommand instead |
-| W001 | warning | Index has orphaned files or dangling references |
-| W002 | warning | MEMORY.md references non-existent topic file |
-| W003 | warning | MEMORY.md exceeds 200 lines -- content truncated at load |
-</error_codes>
-
-<success_criteria>
-- [ ] Store paths resolved correctly for both workflow and system stores
-- [ ] Subcommand parsed and validated (defaults to list)
-- [ ] list: displays entries from selected stores with filtering
-- [ ] search: full-text grep across stores, ranked by match count
-- [ ] view: auto-detects store, displays full content
-- [ ] edit: reads and applies changes to system memory files
-- [ ] delete: requires confirmation, prevents MEMORY.md deletion
-- [ ] prune: requires filter, supports --dry-run, workflow store only
-- [ ] Integrity check after write operations (orphans, dangling refs)
-</success_criteria>
+---
+name: manage-knowhow
+description: Manage knowhow entries across workflow and system stores (list, search, view, edit, delete, prune)
+argument-hint: "[list|search|view|edit|delete|prune] [query|id|file] [--store workflow|system|all] [--tag tag] [--type type]"
+allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+---
+
+<purpose>
+Manage knowhow entries across workflow and system stores. Provides list, search, view, edit, delete, and prune operations over `.workflow/knowhow/` (workflow store) and `~/.claude/projects/{project}/memory/` (system store).
+</purpose>
+
+<context>
+$ARGUMENTS — subcommand followed by options. Defaults to `list` if no arguments.
+
+```bash
+$manage-knowhow
+$manage-knowhow "list --store workflow"
+$manage-knowhow "search authentication"
+$manage-knowhow "view KNW-20260318-001"
+$manage-knowhow "edit MEMORY.md"
+$manage-knowhow "delete TIP-20260318-001 --confirm"
+$manage-knowhow "prune --before 2026-01-01 --type tip --dry-run"
+```
+
+**Subcommands**: `list`, `search`, `view`, `edit`, `delete`, `prune`.
+
+**Flags**:
+- `--store workflow|system|all` — Target store (default: all)
+- `--tag <tag>` — Filter by tag
+- `--type <session|tip|template|recipe|reference|decision>` — Filter by knowhow type
+- `--confirm` — Skip delete confirmation prompt
+- `--before <date>` / `--after <date>` — Date filters for prune
+- `--dry-run` — Preview prune without deleting
+</context>
+
+<execution>
+
+### Step 1: Resolve Store Paths
+
+- **Workflow store**: `.workflow/knowhow/` (entries: `KNW-*.md`, `TIP-*.md`, `TPL-*.md`, `RCP-*.md`, `REF-*.md`, `DCS-*.md`, indexed in `.workflow/wiki-index.json`)
+- **System store**: `~/.claude/projects/{project}/memory/` (files: `MEMORY.md` + topic `.md` files)
+
+Derive system path from project root (replace path separators with `--`, prefix drive letter).
+
+### Step 2: Parse Subcommand
+
+Default to `list` if no arguments. Parse first token as subcommand.
+
+### Step 3: Execute Subcommand
+
+**list**: Show entries from both stores (or filtered by `--store`, `--tag`, `--type`).
+- Workflow: use `maestro wiki list --type knowhow --json` or read `.workflow/wiki-index.json`, display ID, type, category, date, tags, title
+- System: list `.md` files in system memory directory
+
+**search `<query>`**: Full-text grep across both stores. Rank by match count.
+
+**view `<id|file>`**: Auto-detect store from format (`KNW-*/TIP-*/TPL-*/RCP-*/REF-*/DCS-*` = workflow, else system). Display full content.
+
+**edit `<file>`**: Edit a system memory file. Read current content, apply changes. Warn if MEMORY.md exceeds 200 lines (W003).
+
+**delete `<id|file>`**: Require confirmation (or `--confirm` flag). MEMORY.md cannot be deleted (E004). Remove entry file (WikiIndexer auto-updates `.workflow/wiki-index.json` on next access).
+
+**prune**: Requires at least one filter (`--tag`, `--type`, `--before`, `--after`). Workflow store only. `--dry-run` previews without deleting.
+
+### Step 4: Integrity Check
+
+After write operations, verify:
+- No orphaned files without index entries (W001)
+- No dangling index references to missing files (W001)
+- System MEMORY.md references valid topic files (W002)
+</execution>
+
+<error_codes>
+| Code | Severity | Description |
+|------|----------|-------------|
+| E001 | error | No stores found — run `Skill({ skill: "manage-knowhow-capture" })` or create MEMORY.md |
+| E002 | error | Entry ID or filename not found |
+| E003 | error | Prune requires at least one filter flag |
+| E004 | error | Cannot delete MEMORY.md — use `edit` subcommand instead |
+| W001 | warning | Index has orphaned files or dangling references |
+| W002 | warning | MEMORY.md references non-existent topic file |
+| W003 | warning | MEMORY.md exceeds 200 lines — content truncated at load |
+</error_codes>
+
+<success_criteria>
+- [ ] Store paths resolved correctly for both workflow and system stores
+- [ ] Subcommand parsed and validated (defaults to list)
+- [ ] list: displays entries from selected stores with filtering
+- [ ] search: full-text grep across stores, ranked by match count
+- [ ] view: auto-detects store, displays full content
+- [ ] edit: reads and applies changes to system memory files
+- [ ] delete: requires confirmation, prevents MEMORY.md deletion
+- [ ] prune: requires filter, supports --dry-run, workflow store only
+- [ ] Integrity check after write operations (orphans, dangling refs)
+</success_criteria>