Claude Code Plugins

Community-maintained marketplace

Feedback

Portfolio-level project tracker for ID8Labs. Tracks all projects through lifecycle states, enforces stage gates, calculates decay, and coordinates review rituals. The nervous system of the ID8Labs pipeline.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name tracker
description Portfolio-level project tracker for ID8Labs. Tracks all projects through lifecycle states, enforces stage gates, calculates decay, and coordinates review rituals. The nervous system of the ID8Labs pipeline.
version 1.0.0
mcps Memory

ID8TRACKER - Pipeline Nervous System

Purpose

Track all ID8Labs projects from idea capture through exit. Enforce quality gates, calculate activity decay, trigger review rituals, and generate portfolio dashboards.


Lifecycle States

CAPTURED → VALIDATING → VALIDATED → ARCHITECTING → BUILDING → LAUNCHING → GROWING → OPERATING → EXITING → EXITED

Special States

  • ICE - Intentionally frozen, decay paused
  • KILLED - Failed/abandoned with lessons logged (terminal)
  • ARCHIVED - Successfully completed (terminal)

Commands

/tracker status [project-slug]

No argument: Show portfolio dashboard With argument: Show detailed project card

Process:

  1. If no argument, read all project cards from .id8labs/projects/active/
  2. Calculate decay for each project
  3. Generate dashboard using templates/dashboard.md
  4. Highlight any projects in warning (50-79%) or critical (80%+) decay

/tracker new <project-slug> <project-name>

Create a new project in CAPTURED state.

Process:

  1. Generate project card from templates/project-card.md
  2. Set state to CAPTURED, created/last_activity to today
  3. Save to .id8labs/projects/active/{slug}.md
  4. Confirm creation with summary

/tracker update <project-slug> <new-state>

Transition project to new state.

Process:

  1. Load project card
  2. Verify transition is valid (see frameworks/project-states.md)
  3. Check gate requirements (see frameworks/stage-gates.md)
  4. If gate passed:
    • Update state
    • Reset decay (state_entered = today)
    • Log transition in state history
    • Save project card
  5. If gate blocked:
    • List unmet requirements
    • Suggest actions to close gaps

/tracker ice <project-slug> [reason]

Freeze a project. Decay stops.

Process:

  1. Load project card
  2. Record previous state and freeze date
  3. Set state to ICE
  4. Log reason (required if not provided, prompt for it)
  5. Move file to .id8labs/projects/ice/
  6. Confirm freeze

/tracker thaw <project-slug>

Revive a frozen project.

Process:

  1. Load project from .id8labs/projects/ice/
  2. Run revival questions (see frameworks/decay-mechanics.md REVIVAL section)
  3. Restore to previous state
  4. Reset decay timer
  5. Move file to .id8labs/projects/active/
  6. Confirm revival with recommitment

/tracker kill <project-slug> [reason]

Terminate a project permanently.

Process:

  1. Load project card
  2. Prompt for lessons learned if not provided
  3. Set state to KILLED
  4. Log reason and lessons
  5. Move to .id8labs/projects/archive/
  6. Confirm kill with lessons summary

/tracker log <project-slug> <activity>

Log activity to a project. Resets decay timer.

Process:

  1. Load project card
  2. Add activity to activity log with timestamp
  3. Update last_activity to today
  4. Save project card
  5. Confirm log entry

/tracker pulse

Daily 2-minute pulse check.

Process:

  1. Load all active projects
  2. Calculate decay for each
  3. Use rituals/daily-pulse.md format
  4. Show:
    • Any critical (80%+) projects
    • Any warnings (50-79%)
    • Recommended focus for today
    • Quick wins available

/tracker review

Weekly 15-minute review.

Process:

  1. Load all projects (active + ice)
  2. Calculate metrics
  3. Use rituals/weekly-review.md format
  4. Walk through each active project:
    • Progress since last review
    • Blockers
    • Gate readiness
  5. Generate recommendations

/tracker strategy

Monthly 30-minute strategy session.

Process:

  1. Load entire portfolio (active + ice + recent archive)
  2. Use rituals/monthly-strategy.md format
  3. Portfolio health analysis
  4. Stage distribution
  5. Ice box cleanup decisions
  6. Pattern recognition
  7. Next month intentions

/tracker dashboard

Regenerate the DASHBOARD.md file.

Process:

  1. Load all projects
  2. Calculate all metrics
  3. Use templates/dashboard.md
  4. Write to .id8labs/dashboard/DASHBOARD.md
  5. Confirm generation

/tracker gates <project-slug>

Show gate requirements for next transition.

Process:

  1. Load project card
  2. Identify current state and target state
  3. Load requirements from frameworks/stage-gates.md
  4. Show checklist with current completion status

/tracker gate-pass <project-slug> <requirement>

Mark a gate requirement as met.

Process:

  1. Load project card
  2. Add requirement to gates_passed array
  3. Save project card
  4. Show updated gate status

Decay Calculation

decay_percent = (days_since_last_activity / state_max_duration) * 100

Decay Windows by State

State Warning (50%) Critical (80%) Freeze (100%)
CAPTURED 7 days 11 days 14 days
VALIDATING 15 days 24 days 30 days
VALIDATED 10 days 17 days 21 days
ARCHITECTING 7 days 11 days 14 days
BUILDING 45 days 72 days 90 days
LAUNCHING 10 days 17 days 21 days
GROWING 90 days 144 days 180 days
OPERATING - - No decay
EXITING 30 days 48 days 60 days

What Resets Decay

  • Invoking any ID8Labs skill for the project
  • Manual /tracker log entry
  • State transition via /tracker update
  • Completing a gate requirement

Integration with Other Skills

When other ID8Labs skills complete work, they MUST log to tracker:

## Handoff Pattern

After skill completion:
1. Save skill outputs
2. Call: /tracker log {project-slug} "{skill-name}: {summary}"
3. If state transition appropriate, suggest: /tracker update {project-slug} {new-state}

Example Integration Points

Skill Completes Log Message Suggested Transition
scout (BUILD) "Scout: Validation complete - BUILD verdict" VALIDATING → VALIDATED
scout (KILL) "Scout: Validation complete - KILL verdict" → KILLED
architect "Architect: Architecture doc complete" VALIDATED → ARCHITECTING
launch "Launch: Product launched to {channel}" BUILDING → LAUNCHING
growth "Growth: Experiment {name} completed" (no transition, activity log)
ops "Ops: SOP created for {process}" GROWING → OPERATING
exit "Exit: Exit memo drafted" OPERATING → EXITING

Memory MCP Integration

Use Memory MCP to store portfolio-level learnings:

Create entities for:
- Successful patterns (what works)
- Kill post-mortems (what failed)
- Review insights (strategic observations)
- Velocity benchmarks (how long things take)

Query Memory when:

  • Starting a new project (recall similar patterns)
  • Hitting a blocker (check if solved before)
  • Completing a stage (log learnings)

Health Status Indicators

Symbol Status Decay Range Action
🟢 Healthy 0-49% None needed
🟡 Warning 50-79% Surface in pulse
🔴 Critical 80-99% Escalate in review
Frozen 100% Block transitions
❄️ Ice N/A Intentionally paused
⚰️ Killed N/A Terminal
Archived N/A Complete

File Operations

Reading Projects

Active: .id8labs/projects/active/*.md
Ice: .id8labs/projects/ice/*.md
Archive: .id8labs/projects/archive/*.md

Writing Projects

  • Parse YAML frontmatter for structured data
  • Preserve markdown content
  • Update timestamps on every write

Dashboard Location

.id8labs/dashboard/DASHBOARD.md

Error Handling

Error Response
Project not found List available projects, suggest /tracker new
Invalid state transition Explain valid transitions, show current state
Gate blocked List unmet requirements with suggestions
Missing required field Prompt for the field
File write error Report error, suggest manual check

Reminder Configuration

Settings stored in .id8labs/config/settings.yaml:

reminders:
  pulse_frequency: daily    # daily, every-other-day, off
  review_frequency: weekly  # weekly, biweekly, off
  strategy_frequency: monthly

nudge_style: direct         # gentle, direct, aggressive

decay:
  warn_threshold: 50
  critical_threshold: 80
  freeze_threshold: 100

Changelog

v1.0.0 (2025-12-21)

  • Initial release
  • Full lifecycle state machine
  • Decay mechanics with configurable windows
  • Gate verification system
  • Review rituals (daily/weekly/monthly)
  • Dashboard generation
  • Memory MCP integration for learnings