maicongambini/omp-harness-guide/harness-termination-check

Claude Code and agent skill by maicongambini/omp-harness-guide

Trust tier 2
  • community
LLM review pending
Source check pending

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

Verify three-layer completion evidence before claiming done. Use when finishing tasks or passing deliverables. Do NOT use for interim runtime planning (use harness-runtime-feedback).

Install

npx vskill@latest install maicongambini/omp-harness-guide/harness-termination-check

SKILL.md

The same SKILL.md was published earlier as maicongambini/omp-harness-guide/harness-termination-check.

Harness Termination Check

Run before saying "done", "complete", or setting any durable status to passing.

Layer 1 — Static

  • [ ] Static checks from sprint contract or discovered stack pass.
  • [ ] Lint/typecheck commands are either run or explicitly not applicable.
  • [ ] No new warnings vs baseline.
  • [ ] harness-quality-gate ran at the mode this position requires and exited 0 — or every breach

is in observe mode and named in the evidence. A missing, stale or unconfigured report is not a pass; record it as unverified.

  • [ ] Every unavailable metric is listed as a gap, not treated as green.

Layer 2 — Runtime Behavior

  • [ ] Acceptance criteria from sprint contract are observable, not assumed.
  • [ ] Commands were run and output recorded, not inferred or remembered.
  • [ ] Regression checks for previously passing work were run or scoped out with reason.

Layer 3 — System Confirmation

  • [ ] Startup command from harness-startup-path exits 0 or skipped reason is recorded.
  • [ ] Feature state/progress state answer what changed for a new session.
  • [ ] Evidence contains actual command output, not a description.

Fail Protocol

If any layer fails:

  1. Record exact failure: command, exit code, and output. For a metric, record **the metric name,

its value, and its threshold** — a description is not evidence, a number is.

  1. Set task to blocked in the available feature state with reason.
  2. Document in the available progress file Blockers.
  3. Do not claim done. Do not write completed handoff.

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

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/maicongambini/omp-harness-guide/harness-termination-check/badge)](https://verified-skill.com/skills/maicongambini/omp-harness-guide/harness-termination-check)

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