# Definition Of Done Review

> >

- **Type:** Skill
- **Install:** `agentstack add skill-enocgit-sdlc-kit-definition-of-done-review`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [enocgit](https://agentstack.voostack.com/s/enocgit)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [enocgit](https://github.com/enocgit)
- **Source:** https://github.com/enocgit/sdlc-kit/tree/main/skills/definition-of-done-review

## Install

```sh
agentstack add skill-enocgit-sdlc-kit-definition-of-done-review
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# definition-of-done-review

A consistent, team-specific final check. Generic `code-review` and `simplify` catch bugs and
cleanup; this verifies the change is actually *done* by our standard. Run it after them.

## Checklist — read it from AGENTS.md, don't rely on a copy

The **canonical Definition of Done lives in the project's `AGENTS.md`** ("Definition of Done"
section). Read it at run time and review the change against **that exact list**, item by item —
this skill deliberately embeds no copy, so a project that tightens or extends its DoD is
automatically reviewed against its own standard. (The sensitive-areas list is likewise canonical
in `AGENTS.md` → Sensitive areas.)

How to judge the items that need interpretation:

- **Acceptance criteria** — quote each criterion from the PRD/issue and map it to evidence.
- **Contract fidelity** — no undocumented endpoints/fields; types derived from the contract, not
  hand-duplicated; any mismatch is a finding even if tests pass.
- **Tests / observed working** — green suite is necessary, not sufficient: the change must have
  been run and observed (`run`/`verify`), and new logic covered, not just touched.
- **Docs** — `docs/architecture.md` if the system's shape changed; a new ADR if a decision was
  made; the tracker issue reflects reality.
- **Security** — for sensitive-area changes, `security-review` was run with findings resolved and
  `docs/security.md` updated.

## Output

Produce a pass/fail verdict per item with file:line evidence. If anything fails, list the exact
follow-ups and do NOT mark the task ready to merge. This is a gate.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [enocgit](https://github.com/enocgit)
- **Source:** [enocgit/sdlc-kit](https://github.com/enocgit/sdlc-kit)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-enocgit-sdlc-kit-definition-of-done-review
- Seller: https://agentstack.voostack.com/s/enocgit
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.