maestro-flow 0.3.38 → 0.3.40

package/.codex/skills/maestro-milestone-release/SKILL.md

@@ -1,70 +1,70 @@
----
-name: maestro-milestone-release
-description: Version bump, changelog generation, and git tag for a completed milestone. Auto-detects version manifest (package.json, pyproject.toml, Cargo.toml), generates changelog from milestone summaries + git log, creates annotated tag.
-argument-hint: "[<version>] [--bump patch|minor|major] [--dry-run] [--no-tag] [--no-push]"
-allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
----
-
-<purpose>
-Package a completed milestone into a releasable version. Bumps project version in manifest,
-generates CHANGELOG.md entry from phase/milestone summaries and git log, creates annotated
-git tag, optionally pushes to remote. Runs after `/maestro-milestone-complete`.
-</purpose>
-
-<required_reading>
-@~/.maestro/workflows/milestone-release.md
-</required_reading>
-
-<context>
-$ARGUMENTS — optional explicit version and flags.
-
-**Flags:**
-- `<version>` — Explicit version (e.g. `1.2.0`). If omitted, derived from `--bump`.
-- `--bump patch|minor|major` — Semver bump (default: `minor`)
-- `--dry-run` — Compute version + changelog without writing
-- `--no-tag` — Skip git tag creation
-- `--no-push` — Skip `git push --follow-tags`
-
-**Preconditions:**
-- Current milestone completed (audit PASS + milestone-complete run)
-- Working tree clean (no uncommitted changes) unless `--dry-run`
-</context>
-
-<execution>
-Follow '~/.maestro/workflows/milestone-release.md' completely.
-
-**Flow:** Validate preconditions → Resolve version → Collect changes → Generate CHANGELOG →
-Bump manifest → Commit → Tag → Push
-
-**Report:**
-```
-=== RELEASE COMPLETE ===
-Version:   v{previous} → v{new}
-Milestone: {name}
-Tag:       v{new} {pushed|local-only}
-Changelog: {N} entries
-```
-</execution>
-
-<error_codes>
-| Code | Severity | Condition | Recovery |
-|------|----------|-----------|----------|
-| E001 | error | Milestone not completed | Run `$maestro-milestone-complete` first |
-| E002 | error | Audit verdict not PASS | Re-run `$maestro-milestone-audit` |
-| E003 | error | Working tree not clean | Commit or stash changes |
-| E004 | error | Version manifest not found | Add manifest or pass explicit version |
-| E005 | error | Version not greater than current | Choose higher version |
-| W001 | warning | No changes since last tag | Confirm release is desired |
-| W002 | warning | Remote push failed | Retry `git push --follow-tags` |
-</error_codes>
-
-<success_criteria>
-- [ ] Preconditions validated
-- [ ] Target version computed and greater than previous
-- [ ] Version manifest(s) updated
-- [ ] CHANGELOG.md entry with milestone summary + grouped changes
-- [ ] Release commit created
-- [ ] Annotated git tag created (unless --no-tag)
-- [ ] Pushed to remote (unless --no-push)
-- [ ] state.json updated with last_release_version
-</success_criteria>
+---
+name: maestro-milestone-release
+description: Bump version, generate changelog, tag milestone
+argument-hint: "[<version>] [--bump patch|minor|major] [--dry-run] [--no-tag] [--no-push]"
+allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+---
+
+<purpose>
+Package a completed milestone into a releasable version. Bumps project version in manifest,
+generates CHANGELOG.md entry from phase/milestone summaries and git log, creates annotated
+git tag, optionally pushes to remote. Runs after `/maestro-milestone-complete`.
+</purpose>
+
+<required_reading>
+@~/.maestro/workflows/milestone-release.md
+</required_reading>
+
+<context>
+$ARGUMENTS — optional explicit version and flags.
+
+**Flags:**
+- `<version>` — Explicit version (e.g. `1.2.0`). If omitted, derived from `--bump`.
+- `--bump patch|minor|major` — Semver bump (default: `minor`)
+- `--dry-run` — Compute version + changelog without writing
+- `--no-tag` — Skip git tag creation
+- `--no-push` — Skip `git push --follow-tags`
+
+**Preconditions:**
+- Current milestone completed (audit PASS + milestone-complete run)
+- Working tree clean (no uncommitted changes) unless `--dry-run`
+</context>
+
+<execution>
+Follow '~/.maestro/workflows/milestone-release.md' completely.
+
+**Flow:** Validate preconditions → Resolve version → Collect changes → Generate CHANGELOG →
+Bump manifest → Commit → Tag → Push
+
+**Report:**
+```
+=== RELEASE COMPLETE ===
+Version:   v{previous} → v{new}
+Milestone: {name}
+Tag:       v{new} {pushed|local-only}
+Changelog: {N} entries
+```
+</execution>
+
+<error_codes>
+| Code | Severity | Condition | Recovery |
+|------|----------|-----------|----------|
+| E001 | error | Milestone not completed | Run `$maestro-milestone-complete` first |
+| E002 | error | Audit verdict not PASS | Re-run `$maestro-milestone-audit` |
+| E003 | error | Working tree not clean | Commit or stash changes |
+| E004 | error | Version manifest not found | Add manifest or pass explicit version |
+| E005 | error | Version not greater than current | Choose higher version |
+| W001 | warning | No changes since last tag | Confirm release is desired |
+| W002 | warning | Remote push failed | Retry `git push --follow-tags` |
+</error_codes>
+
+<success_criteria>
+- [ ] Preconditions validated
+- [ ] Target version computed and greater than previous
+- [ ] Version manifest(s) updated
+- [ ] CHANGELOG.md entry with milestone summary + grouped changes
+- [ ] Release commit created
+- [ ] Annotated git tag created (unless --no-tag)
+- [ ] Pushed to remote (unless --no-push)
+- [ ] state.json updated with last_release_version
+</success_criteria>

package/.codex/skills/maestro-overlay/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: maestro-overlay
-description: Create or edit a non-invasive command overlay from natural-language intent. Writes a JSON patch file to ~/.maestro/overlays/, applies it via maestro overlay add, and confirms installation with idempotent re-apply support.
+description: Create or edit command overlays from natural language
 argument-hint: "<intent> | --list | --remove <name>"
 allowed-tools: Read, Write, Bash, Glob, Grep
 ---

package/.codex/skills/maestro-plan/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: maestro-plan
-description: Exploration-driven planning via CSV wave pipeline. Wave 1 runs parallel codebase exploration agents, Wave 2 synthesizes explorations into plan.json + TASK-*.json. Replaces maestro-plan command.
+description: Plan phase execution with exploration and verification
 argument-hint: "[-y|--yes] [-c|--concurrency N] [--continue] \"<phase> [--dir <path>] [--gaps] [--spec SPEC-xxx] [--collab]\""
-allowed-tools: spawn_agents_on_csv, Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+allowed-tools: spawn_agents_on_csv, Read, Write, Edit, Bash, Glob, Grep, request_user_input
 ---
 
 <purpose>
@@ -46,6 +46,7 @@ Wave-based planning using `spawn_agents_on_csv`. Wave 1 explores codebase contex
 |                                                                           |
 +---------------------------------------------------------------------------+
 ```
+
 </purpose>
 
 <context>
@@ -168,6 +169,20 @@ Each wave generates `wave-{N}.csv` with extra `prev_context` column.
 // Create: sessionFolder, scratchDir/.task/
 ```
 
+### Session Resume (`--continue`)
+
+When `continueMode` is true:
+1. Scan `.workflow/.csv-wave/` for directories matching `*-plan-*`
+2. If `$ARGUMENTS` contains a session ID suffix, match it; otherwise use the most recent session
+3. Read the session's `tasks.csv` to determine resume point:
+   - If all wave 1 tasks completed → skip to wave 2
+   - If wave 2 completed → skip to Phase 3 (plan checking)
+   - If Phase 3 incomplete → re-run plan checking
+4. Re-use existing `sessionFolder` and `scratchDir` (do not create new ones)
+5. Log: "Resuming session {sessionId} — skipping {N} completed tasks"
+
+If no matching session found, list available sessions and abort.
+
 ### Phase 1: Phase Resolution -> CSV
 
 **Objective**: Resolve phase, load context, determine exploration angles, generate tasks.csv.
@@ -259,7 +274,7 @@ ELSE:
 spawn_agents_on_csv({
   csv_path: `${sessionFolder}/wave-1.csv`,
   id_column: "id",
-  instruction: buildExplorationInstruction(sessionFolder, phaseDir),
+  instruction: buildExplorationInstruction(sessionFolder, phaseDir),  // agent: ~/.codex/agents/cli-explore-agent.toml
   max_concurrency: maxConcurrency,
   max_runtime_seconds: 3600,
   output_csv_path: `${sessionFolder}/wave-1-results.csv`,
@@ -298,7 +313,7 @@ Include this ceiling in `buildPlanningInstruction`. If the planning agent output
 spawn_agents_on_csv({
   csv_path: `${sessionFolder}/wave-2.csv`,
   id_column: "id",
-  instruction: buildPlanningInstruction(sessionFolder, phaseDir, {
+  instruction: buildPlanningInstruction(sessionFolder, phaseDir, {  // agent: ~/.codex/agents/workflow-planner.toml
     contextMd, indexJson, specRef, docIndex, designRef, gapsContext
   }),
   max_concurrency: 1,  // Single planning agent
@@ -344,6 +359,12 @@ spawn_agents_on_csv({
 1. **Plan checking** (inline, not a separate wave):
    Read `plan.json` + all `.task/TASK-*.json`. Validate: requirements coverage, file feasibility, dependency correctness (no cycles, valid wave order), grep-verifiable convergence criteria, read_first completeness, action concreteness, no parallel file conflicts, **task count within complexity threshold** (reject over-split plans), **no per-file splitting** (each task must be feature-level).
 
+1b. **Plan confidence scoring**:
+
+   Dimensions (5): requirements_coverage, task_quality, dependency_correctness, estimation_accuracy, collision_safety. Factors (weights): completeness(.30), specificity(.25), structural_validity(.20), user_validation(.15), consistency(.10). Add `confidence` section to `plan.json`.
+
+   **Readiness gate**: Block if requirements_coverage < 40% or any task missing read_first/convergence.criteria.
+
 2. **Revision loop** (max 3 rounds): If critical issues found, regenerate affected tasks.
 
 2b. **Spec Enrichment**: Persist cross-task reusable design decisions:
@@ -382,6 +403,7 @@ spawn_agents_on_csv({
    Tasks: {task_count} tasks in {wave_count} waves
    Check: {checker_status} (iteration {check_count}/{max_checks})
    Collision: {collision_status}
+   Confidence: {overall}% (weakest: {dim})
 
    Next steps:
      $maestro-execute "{phase}"     -- Execute the plan
@@ -449,6 +471,9 @@ echo '{"ts":"<ISO>","worker":"{id}","type":"existing_pattern","data":{"name":"Re
 - [ ] plan.json produced in phase directory
 - [ ] .task/TASK-*.json files produced for all tasks
 - [ ] Plan passes quality checks (coverage, deps, criteria)
+- [ ] Plan confidence scored with 5-dimension factor model
+- [ ] Readiness gate checked before confirmation
+- [ ] plan.json includes confidence section
 - [ ] Collision detection executed against same-milestone plans
 - [ ] PLN artifact registered in state.json
 - [ ] context.md produced with exploration findings + plan overview