lennney/ticketpilot · Archived

project-plan-audit

Use when auditing TicketPilot project plans, modules, tests, OpenSpec changes, and phase status before continuing development.

First seen Jun 21, 2026

Installation

$ npx skills add lennney/ticketpilot --skill project-plan-audit

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from lennney/ticketpilot.

npx skills add lennney/ticketpilot

Browse all from lennney/ticketpilot

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch master
Open issues 3
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,955 B
  • docs SUMMARY.md 152 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Project Plan Audit Skill

The audit must include:

1. Scope Alignment

Check whether the project still follows the TicketPilot MVP scope:

  • ticket normalization
  • intent classification
  • risk gate
  • FAQ / Policy / Case layered retrieval
  • hybrid retrieval (keyword + vector + RRF)
  • evidence-grounded draft reply
  • human review
  • trace
  • evaluation

Mark any generic chatbot, unnecessary multi-agent product feature, or over-engineered component as scope drift.

2. Requirement-to-Evidence Matrix

For each major requirement, map:

  • requirement
  • OpenSpec source
  • implementation file
  • test file / test command
  • status: PASS / GAP / FAIL
  • notes

3. Module Boundary Review

For each module, check:

  • purpose
  • input
  • output
  • dependency
  • test coverage
  • whether it is too broad or too coupled

4. Test Coverage Review

Review:

  • unit tests
  • integration tests
  • skipped tests
  • smoke tests
  • evaluation cases
  • missing negative cases

Rule: A skipped integration test is NOT a pass. It is a GAP.

Rule: Fake embedding proves pipeline wiring only, not semantic retrieval quality.

5. Phase Status Accuracy

Check:

  • docs/phase_status.md
  • docs/changelog.md
  • README.md
  • docs/technical_decisions.md

Ensure status claims are not overstated.

6. Risk Register

Classify issues:

  • BLOCKING
  • HIGH
  • MEDIUM
  • LOW
  • DEFERRED

7. Minimal Fix Plan

Produce a prioritized fix plan:

  • Fix now
  • Fix before next feature
  • Defer safely
  • Remove / simplify

Do not recommend broad rewrites unless absolutely necessary. Do not propose new features during audit.

8. Final Decision

Output one of:

  • CONTINUE — all clear to proceed with next phase
  • CONTINUEWITHGAPS — proceed but track gaps explicitly
  • HOLDNEWFEATURES — blockers must be resolved first