rafaie/codex_home/feature-kickoff

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Create and initialize a complete work item documentation folder from a backlog item or clarified brief. Use before spec-linter and implementation-phase.

Install

npx vskill@latest install rafaie/codex_home/feature-kickoff

SKILL.md

1) Read project AGENTS.md, codex.toml, spec/brief.md, and the relevant spec/backlog.md item when present. 2) Collect inputs: Work item ID (S-<stream>-<nnn>, example S-core-001), title, goal, constraints.

  • Accept legacy dotted F IDs only when updating older specs; create new work with S-<stream>-<nnn>.

3) Create a work item folder spec/features/<work-id>-<slug>/ (use actual work item ID).

  • Prefer repo templates under spec/templates/ if present.
  • Otherwise use this skill's assets/default_*_template.md files.

4) Create/update these documents inside the folder:

  • feature.md: problem, scope, acceptance criteria, design, ADR links
  • implementation.md: implementation plan and chronological implementation log
  • test-plan.md: unit/integration/e2e/edge/smoke test plan
  • test-results.md: commands run, outcomes, failures, fixes
  • status.md: status, owner, next action, blockers
  • evidence/README.md: directory placeholder and evidence index

5) Fill the initial docs.

  • Acceptance criteria must use stable IDs such as AC-1 and begin unchecked.
  • test-plan.md must set smoke applicability to required or not_applicable.
  • Required smoke must include scenario, fixture, pass/fail mode, runtime budget, and artifact path; not-applicable smoke must explain why the work is non-runnable.

6) Create an ADR only when a concrete architecture/design decision is present; never create an ADR solely because spec/decisions/ is empty.

  • Use spec/decisions/ADR-XXXX-<slug>.md and prefer spec/templates/adr.md.

7) Stop after writing docs. Do NOT implement code. 8) End by listing the next 3 tasks.

  • Tasks must be distinct (no duplicates).
  • Use this default sequence unless the user asks for a different flow:

1) Run $spec-linter <work-id> to review the work item docs for gaps. 2) Run $implementation-phase <work-id> to plan tests, write tests, and implement the work item. 3) Run $ship-feature <work-id> to execute full checks, smoke evidence, ADR/docs/index updates, and final readiness output.

  • <work-id> must be the real ID for the folder you just created (example: S-core-001).
  • Keep each task line concrete and non-redundant.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/rafaie/codex_home/feature-kickoff/badge)](https://verified-skill.com/skills/rafaie/codex_home/feature-kickoff)

Something wrong with this skill or its page? Tell us.