Navigator Task Manager Skill

Create and manage task documentation - implementation plans that capture what was built, how, and why.

When to Invoke

Invoke this skill when the user:

• Says "document this feature", "archive this task"
• Says "create task doc for...", "document what I built"
• Completes a feature and mentions "done", "finished", "complete"
• Starts new feature and says "create implementation plan"

DO NOT invoke if:

• User is asking about existing tasks (use Read, not creation)
• Creating SOPs (that's nav-sop skill)
• Updating system docs (different skill)

Execution Steps

Step 1: Determine Task ID

If user provided task ID (e.g., "TASK-01", "GH-123"):

• Use their ID directly

If no ID provided, run the generator — it reads task_id_source from .agent/.nav-config.json (default local):

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$(cat "${NAVIGATOR_CONFIG_HOME:-${XDG_CONFIG_HOME:-$HOME/.config}/navigator}/plugin-root" 2>/dev/null)}"
[ -d "$PLUGIN_DIR/skills" ] || PLUGIN_DIR="$HOME/.claude/plugins/marketplaces/navigator-marketplace"
python3 "$PLUGIN_DIR/skills/nav-task/functions/task_id_generator.py" \
  --title "{feature name}" --body "{one-line summary}" --json

• task_id_source: local (solo repos) — next sequential {task_prefix}-NN from .agent/tasks/ (last is TASK-05 → TASK-06). --title is ignored. The number is the max over the working tree, tasks/archive/ and origin/main (fetched first; a failing fetch falls back to the local two with a note). Open remote branches are not scanned: when two sessions plan at once, the second to push renumbers.
• task_id_source: github (team repos, GH-32) — creates the GitHub issue first via gh issue create and returns GH-<n>; the doc is .agent/tasks/GH-<n>-{slug}.md. Issue numbers are allocated by GitHub, so two contributors branching at once can never mint the same ID, and Pilot already addresses tasks as GH-<n>. Add --label pilot (or any label) when the issue should be picked up. A failing gh (no auth, no network) is an error — do not fall back to a local number; tell the user and stop.
• Existing TASK-NN docs keep working either way; the index, graph sync and session-start listing accept any PREFIX-<n> filename.
• In the doc template below, TASK-{XX} stands for whichever ID came back.

Step 2: Determine Action (Create vs Archive)

Creating new task (starting feature):

User: "Create task doc for OAuth implementation"
→ Action: CREATE
→ Generate empty implementation plan template

Archiving completed task (feature done):

User: "Document this OAuth feature I just built"
→ Action: ARCHIVE
→ Generate implementation plan from conversation

Step 2.5: Recall Prior Knowledge (CREATE flow only, v6.17.0+)

Before writing the plan, query the knowledge graph for memories relevant to this feature — pitfalls and patterns the project has already paid for:

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$(cat "${NAVIGATOR_CONFIG_HOME:-${XDG_CONFIG_HOME:-$HOME/.config}/navigator}/plugin-root" 2>/dev/null)}"
[ -d "$PLUGIN_DIR/skills" ] || PLUGIN_DIR="$HOME/.claude/plugins/marketplaces/navigator-marketplace"

if [ -f ".agent/knowledge/graph.json" ]; then
  python3 "$PLUGIN_DIR/skills/nav-graph/functions/memory_recall.py" \
    --concepts "{comma-separated concepts inferred from the feature description}" \
    --format markdown --limit 5
fi

• Infer concepts from the feature description using the same keyword families as task_to_graph.extract_concepts_from_task (auth, database, api, frontend, testing, …) — alias resolution absorbs abbreviations.
• Empty output → skip silently. Do not add the section; do not mention the absence.
• Non-empty output → include it as the ## Known Pitfalls & Patterns section in the template below, and reflect each recalled pitfall in the Implementation phases (a recalled pitfall that doesn't change the plan wasn't really factored in).

Step 3A: Create New Task (If Starting Feature)

Generate task document from template:

# TASK-{XX}: {Feature Name}

**Status**: 🚧 In Progress
**Created**: {YYYY-MM-DD}
**Assignee**: {from PM tool or "Manual"}

---

## Context

**Problem**:
[What problem does this solve?]

**Goal**:
[What are we building?]

---

## Known Pitfalls & Patterns

<!-- From knowledge graph (Step 2.5). OMIT this section entirely if recall
     returned nothing. Each recalled pitfall must be reflected in the
     Implementation phases below. -->

- **PITFALL** (90%, mem-XXX): [recalled summary]

---

## Acceptance Criteria

Concrete, checkable outcomes — written so anyone (human or AI) can verify them.

- [ ] [Specific, observable outcome]
- [ ] [Another outcome]
- [ ] [Edge case handled]

---

## Implementation

### Phase 1: {Name}
**Goal**: [What this phase accomplishes]

**Tasks**:
- [ ] [Specific task]
- [ ] [Another task]

**Files**:
- `path/to/file.ts` - [Purpose]

### Phase 2: {Name}
...

---

## Out of Scope

Explicit non-goals — what this task deliberately does not address.

- [What's deferred to a future task]
- [Adjacent change being avoided]

---

## Technical Decisions

| Decision | Options Considered | Chosen | Reasoning |
|----------|-------------------|--------|-----------|
| [What] | [Option A, B, C] | [Chosen] | [Why] |

---

## Verify

Run these commands to validate the implementation:

```bash
# Run tests
[test command for this feature]

# Type check
[type check command]

# Build
[build command]

Done

Observable outcomes that prove completion:

• [Specific file/API exists and exports expected interface]
• [Tests pass - specify count or coverage target]
• [Build succeeds without errors]
• [User-observable behavior works as specified]

Refs

• [Source plan / design doc / parent issue]
• [Related ticket]

Notes

[Any additional context, links, references]

Last Updated: {YYYY-MM-DD}

Save to: `.agent/tasks/TASK-{XX}-{slug}.md`

### Step 3B: Archive Completed Task (If Feature Done)

Generate task document from conversation:

1. **Analyze conversation** (last 30-50 messages):
   - What was built?
   - How was it implemented?
   - What decisions were made?
   - What files were modified?

2. **Generate implementation plan**:

```markdown
# TASK-{XX}: {Feature Name}

**Status**: ✅ Completed
**Created**: {YYYY-MM-DD}
**Completed**: {YYYY-MM-DD}

---

## What Was Built

[1-2 paragraph summary of the feature]

---

## Implementation

### Phase 1: {Actual phase completed}
**Completed**: {Date}

**Changes**:
- Created `src/auth/oauth.ts` - OAuth provider integration
- Modified `src/routes/auth.ts` - Added login/logout endpoints
- Updated `src/config/passport.ts` - Passport configuration

**Key Code**:
```typescript
// Example of key implementation
export const oauthLogin = async (req, res) => {
  // Implementation details
};

Phase 2: {Next phase}

...

Technical Decisions

Files Modified

• src/auth/oauth.ts (created) - OAuth integration
• src/routes/auth.ts (modified) - Added auth endpoints
• src/config/passport.ts (created) - Passport setup
• tests/auth.test.ts (created) - Auth tests
• README.md (updated) - OAuth setup instructions

Challenges & Solutions

Challenge: OAuth callback URL mismatch

• Problem: Redirects failed in production
• Solution: Added environment-specific callback URLs
• Commit: abc1234

Challenge: Session persistence across restarts

• Problem: Users logged out on server restart
• Solution: Redis session store
• Commit: def5678

Testing

• ✅ Unit tests: src/auth/*.test.ts (15 tests, 100% coverage)
• ✅ Integration tests: OAuth flow end-to-end
• ✅ Manual testing: Tested with Google, GitHub providers

Documentation

• ✅ README updated with OAuth setup instructions
• ✅ Environment variables documented in .env.example
• ✅ API endpoints documented in docs/api.md

Verify

Commands executed to validate:

# Actual commands run during verification
npm test src/auth
npm run type-check
npm run build

Results: All passed ✅

Done

Outcomes confirmed:

• src/auth/oauth.ts exports OAuth provider integration
• All tests pass (15 tests, 100% coverage)
• Build succeeds without errors
• OAuth login/logout flows work correctly

Related

SOPs Created:

• .agent/sops/integrations/oauth-setup.md

System Docs Updated:

• .agent/system/project-architecture.md (added auth section)

Completed: {YYYY-MM-DD} Implementation Time: {X hours/days}

Save to: `.agent/tasks/TASK-{XX}-{slug}.md`

### Step 3.5: Verify Interpretation (ToM Checkpoint - Archive Mode Only) [EXECUTE]

**IMPORTANT**: This step MUST be executed when archiving tasks (not creating new ones).

**Before committing archive documentation, confirm interpretation with user**.

**Display verification** (only for ARCHIVE action):

I extracted this from our session: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

What was built:

• {FEATURE_SUMMARY}

Key decisions captured:

• {DECISION_1}: {REASONING_1}
• {DECISION_2}: {REASONING_2}

Files changed: {COUNT} total

• {FILE_1} ({ACTION}: {PURPOSE})
• {FILE_2} ({ACTION}: {PURPOSE})

Challenges solved:

• {CHALLENGE_1}: {SOLUTION_1} ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Corrections needed? [Enter to proceed / type corrections]

**Always verify for ARCHIVE** because:
- Extracting from conversation is inference-based
- User may have made decisions not explicitly stated
- Some files may have been modified outside conversation
- Ensures accurate historical record

**Skip verification for CREATE** because:
- Template is mostly empty
- User fills in details themselves
- No inference risk

### Step 4: Update Navigator Index

Edit `.agent/DEVELOPMENT-README.md` to add task to index:

```markdown
## Active Tasks

- **TASK-{XX}**: {Feature Name} (Status: In Progress/Completed)
  - File: `.agent/tasks/TASK-{XX}-{slug}.md`
  - Started: {Date}
  - [Completed: {Date}]

Keep index organized (active tasks first, completed below).

Step 4.5: Sync to Knowledge Graph (v6.0.0+)

If knowledge graph exists, sync task to graph:

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$(cat "${NAVIGATOR_CONFIG_HOME:-${XDG_CONFIG_HOME:-$HOME/.config}/navigator}/plugin-root" 2>/dev/null)}"
[ -d "$PLUGIN_DIR/skills" ] || PLUGIN_DIR="$HOME/.claude/plugins/marketplaces/navigator-marketplace"
if [ -f ".agent/knowledge/graph.json" ]; then
  python3 "$PLUGIN_DIR/skills/nav-graph/functions/task_to_graph.py" \
    --action add \
    --task-path ".agent/tasks/TASK-{XX}-{slug}.md" \
    --graph-path .agent/knowledge/graph.json
fi

What this does:

• Extracts concepts from task content (auth, api, testing, etc.)
• Adds task node to graph with status and concepts
• Creates implements edges from task to concepts
• For completed tasks: Extracts Technical Decisions as decision memories

Output:

Added task: TASK-XX
Title: {Feature Name}
Status: completed
Concepts: auth, api, testing
Decisions extracted: 2

This makes the task queryable via "What do we know about auth?" and preserves architectural decisions as persistent memories.

Step 5: Update PM Tool (If Configured)

If PM tool is Linear:

create_comment({
  issueId: "TASK-XX",
  body: "📚 Implementation plan documented: .agent/tasks/TASK-XX-feature.md"
})

If PM tool is GitHub:

gh issue comment {ISSUE-NUMBER} -b "📚 Implementation plan: .agent/tasks/TASK-XX-feature.md"

If PM tool is none: Skip PM update.

Step 6: Confirm Success

Show completion message:

✅ Task documentation created!

Task: TASK-{XX} - {Feature Name}
File: .agent/tasks/TASK-{XX}-{slug}.md
Size: {X} KB (~{Y} tokens)

📋 Contains:
- Implementation phases
- Technical decisions
- Files modified
- [If archived: Challenges & solutions]
- [If archived: Testing & documentation]

🔗 Navigator index updated
[If PM tool: PM tool comment added]

To reference later:
Read .agent/tasks/TASK-{XX}-{slug}.md

Task Document Template Structure

For New Tasks (Planning)

1. Context (problem/goal)
2. Acceptance criteria (- [ ] checkable outcomes)
3. Implementation (phases)
4. Out of scope (explicit non-goals)
5. Technical decisions (to be made)
6. Verify + Done (validation + completion)
7. Refs (sources/related)

For Completed Tasks (Archive)

1. What was built (summary)
2. Implementation (actual phases)
3. Technical decisions (what was chosen)
4. Files modified
5. Challenges & solutions
6. Testing & documentation

Common Use Cases

Starting New Feature

User: "Create task doc for payments integration"
→ Generates TASK-07-payments.md
→ Empty template for planning
→ User fills in as they work

Completing Feature

User: "Document the auth feature I just finished"
→ Analyzes conversation
→ Generates TASK-06-auth.md
→ Complete implementation record
→ Archives for future reference

Mid-Feature Update

User: "Update TASK-05 with OAuth decision"
→ Reads existing TASK-05-auth.md
→ Adds to Technical Decisions section
→ Preserves rest of document

Error Handling

Navigator not initialized:

❌ .agent/tasks/ directory not found

Run /nav:init to set up Navigator structure first.

Task ID already exists (for creation):

⚠️  TASK-{XX} already exists

Options:
1. Read existing task
2. Use different ID
3. Archive/overwrite existing

Your choice [1-3]:

Insufficient context to archive:

⚠️  Not enough conversation context to generate implementation plan

Consider:
- Provide more details about what was built
- Manually create task doc
- Skip archiving

Continue with template? [y/N]:

Success Criteria

Task documentation is successful when:

• Task file created in .agent/tasks/
• Filename follows convention: TASK-{XX}-{slug}.md
• Contains all required sections
• Navigator index updated
• PM tool updated (if configured)
• User can reference task later

Scripts

generate_task.py: Create task documentation from conversation

• Input: Conversation history, task ID
• Output: Formatted task markdown

update_index.py: Update DEVELOPMENT-README.md task index

• Input: New task info
• Output: Updated index

Best Practices

Good task slugs:

• oauth-implementation (descriptive)
• stripe-payment-flow (clear purpose)
• user-profile-page (specific feature)

Bad task slugs:

• feature (too vague)
• fix (not descriptive)
• task1 (meaningless)

When to create task docs:

• ✅ Starting major feature (> 1 day work)
• ✅ Completing any feature (archive)
• ✅ Complex implementation (capture decisions)
• ❌ Tiny bug fixes (use SOPs instead)
• ❌ Exploratory work (wait until direction clear)

Notes

Task docs are living documents:

• Created when starting feature (template)
• Updated during implementation (decisions)
• Finalized when complete (archive)

They serve as:

• Planning tool (before implementation)
• Progress tracker (during implementation)
• Historical record (after completion)
• Knowledge base (for team/future)

This skill provides same functionality as /nav:doc feature command but with natural language invocation.