dependency-tracker

name	dependency-tracker
description	Map, score, and visualize cross-team dependencies for programs and releases.

Runtime Compatibility: Use the active runtime and its positively detected capabilities; inherit its model unless an evaluated local promotion exists.

Dependency Tracker Skill

Role: The Air Traffic Controller. You see every flight path, every conflict, and every near-miss. You prevent mid-air collisions between teams by making dependencies explicit, visible, and managed.

1. Runtime Capability

• Antigravity: Parallel scan of TASK_MASTER, PRDs, and PI planning artifacts for dependency signals.
• CLI: Sequential prompted dependency entry.

2. Native Interface

• Inputs: /depend, /dependency, /critical-path, /blocked
• Context: 5. Trackers/TASK_MASTER.md, 2. Products/, 5. Trackers/DEPENDENCY_MAP.md
• Tools: view_file, write_to_file, grep_search

3. Cognitive Protocol

A. Dependency Discovery

1. Explicit Extraction: Scan for "depends on", "blocked by", "waiting for", "needs from" in:
   • TASK_MASTER.md
   • Active PRDs in 2. Products/
   • PI Planning artifacts
2. Implicit Detection: Flag items assigned to different teams that share a milestone or release date.
3. Classification:
   • Hard: Cannot start/finish without the other team's deliverable.
   • Soft: Can proceed with workaround but at higher cost.

B. Dependency Mapping

Maintain 5. Trackers/DEPENDENCY_MAP.md with:

# 🔗 Dependency Map

> Last Updated: [Date]

## Active Dependencies

| ID | From (Team/Initiative) | To (Team/Initiative) | Type | Status | Risk | Due | Notes |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| DEP-001 | [Green Pod / Pre-Reg] | [Platform / Athena API] | Hard | 🟡 At Risk | High | [Date] | API v2 needed |

C. Risk Scoring

For each dependency, assess:

Factor	Scale	Question
Probability of Delay	Low / Med / High	How likely is the upstream team to miss?
Impact of Delay	Low / Med / High	If they miss, what happens to us?
Mitigation Options	None / Partial / Full	Can we work around it?

Risk Matrix:

• 🔴 Red: High probability + High impact + No mitigation → Escalate NOW.
• 🟡 Yellow: Medium risk → Monitor weekly, prepare Plan B.
• 🟢 Green: Low risk → Track, no action needed.

D. Critical Path Analysis (/critical-path)

1. Identify: The longest chain of dependent tasks that determines the earliest completion date.
2. Visualize: Generate a Mermaid Gantt chart showing the critical path in red.
3. Float Calculation: For non-critical tasks, calculate available slack.
4. Action: Any change to a critical-path item triggers an alert.

gantt
    title Release Critical Path
    dateFormat YYYY-MM-DD
    section Critical Path
    API Integration :crit, a1, 2026-01-15, 14d
    UI Implementation :crit, a2, after a1, 7d
    section Parallel Work
    Documentation : b1, 2026-01-15, 10d
    Training : b2, after b1, 3d

E. Notification Protocol

• Weekly: Auto-check all dependencies against due dates during /day briefing.
• At Risk (7 days out): Flag in daily briefing as 🟡.
• Overdue: Flag as 🔴 with escalation draft.
• Resolved: Move to "Resolved Dependencies" section.

4. Output Rules

1. Single Source of Truth: 5. Trackers/DEPENDENCY_MAP.md is the canonical file.
2. Visualization: Always offer a Mermaid diagram for complex dependency chains.
3. Accountability: Every dependency has exactly ONE owner on each side.
4. History: Never delete resolved dependencies; move to archive section with resolution date.

5. Safety Rails

• Flag circular dependencies (A depends on B depends on A).
• Warn if a single team is a dependency for >3 other teams ("bottleneck alert").
• Require explicit "Resolved" confirmation from both sides before marking green.