# Assess Against Mission

> Judges whether the work achieves the target.

- Skill: `elliottlawson/assess-against-mission` (Agent Skill)
- Install (CLI): `npx skillmds@latest add elliottlawson/assess-against-mission`
- Raw SKILL.md: https://api.skillmd.com/api/skills/elliottlawson/assess-against-mission/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: elliottlawson (https://skillmd.com/u/elliottlawson)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/elliottlawson/assess-against-mission

---


# Assess against mission

Judge whether the work achieves the target, or does something else.

## Findings

| Kind | Meaning |
|------|---------|
| **missing** | The target required it; the change omits it or only partially delivers |
| **wrong** | Aimed at a requirement but implements it incorrectly |
| **creep** | Behavior the target did not ask for |

Quote or paraphrase the target for each finding. Missing and wrong are
blocking; creep is advisory unless the extra behavior is material.

## Output

- **Met** — the work achieves the target.
- **Missing** — the work falls short; name what's missing.
- **Different** — the work does something other than the target.
- **Unclear** — no target found after asking once. The work might hit the target or miss it — don't grade against an assumed one.