AGENT SKILL

verify-before-done

Use before claiming any task is complete, fixed, passing, or ready. Triggers on: 'done', 'fixed', 'tests pass', 'should work now', 'looks good', 'complete', 'finished', 'ready to merge', or any expression of satisfaction before showing evidence.

License 许可
Apache-2.0
最近更新
2026-11-14
Stars
★ 13
Forks
2

01 前言 Front Matter

name
verify-before-done
description
Use before claiming any task is complete, fixed, passing, or ready. Triggers on: 'done', 'fixed', 'tests pass', 'should work now', 'looks good', 'complete', 'finished', 'ready to merge', or any expression of satisfaction before showing evidence.

03 SKILL.md 正文

Verify Before Done Skill

A claim without evidence is a guess. This skill enforces evidence before assertion.

This is the skill-layer complement to the Truth Gate hook (core/hooks/truth-gate-guard.sh). The hook warns on claim verbs in output. This skill runs the actual verification.

The Rule

NO COMPLETION CLAIM WITHOUT FRESH VERIFICATION OUTPUT IN THIS SAME RESPONSE.

If the verification command was not run in this message, you cannot claim it passes.

When to use this skill

ALWAYS before any of these:

  • Saying "done", "fixed", "complete", "finished", "ready"
  • Expressing satisfaction: "looks good", "should work", "that should do it"
  • Committing or pushing
  • Moving to the next task
  • Telling the user the bug is resolved

Verification checklist

Run the relevant checks and show their output:

Tests

# Run and show full output — do not truncate
bash core/tests/hooks/run-hook-tests.sh 2>&1

Claim "tests pass" only if output shows 0 failed.

Hook syntax

bash -n core/hooks/*.sh 2>&1 && echo "syntax OK"

Claim "syntax clean" only if output shows "syntax OK".

Git state

git diff --stat
git status --short

Show actual output. "No uncommitted changes" requires empty output.

For bug fixes — reproduce before claiming fixed

# Run the specific command/test that was failing
# Show it passes now
# If possible, show it would have failed before (git stash → run → git stash pop)

Output format

After running checks, report like this:

Verification:
- tests: [N/N pass | FAIL: list failures]
- syntax: [OK | errors: list]
- git: [clean | N files changed: list]
- bug: [reproduced and resolved | not verified]

Status: DONE ✓   (or)   NOT DONE — [what's still failing]

Red flags — stop and verify before continuing

  • You are about to type "done", "fixed", "complete"
  • You feel satisfied with the result
  • The last thing you did was edit a file
  • You're about to switch to a new task
  • You're about to run git commit

Any of the above = run this skill first.

Common rationalisations — ignore them

What you're thinking What to do instead
"I just changed one line, clearly it works" Run the test
"The test passed before, it still passes" Run it fresh
"Agent said success" Verify independently
"I'm confident" Confidence ≠ evidence
"Just this once" No exceptions

04 同仓库更多 Skills

Horonomos-laps
Horonomos-laps/Yana-AI

openai--sentry--sentry

Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

ito-basket-compare

Compare Itô prediction-market baskets against a user's knowledge base, portfolio notes, financial context, watchlist, or research thesis. Use for read-only basket comparison and gap analysis without investment advice or live trading.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

implementing-network-traffic-baselining

Build network traffic baselines from NetFlow/IPFIX data using Python pandas for statistical analysis, z-score anomaly detection, and hourly/daily traffic pattern profiling

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

detecting-evasion-techniques-in-endpoint-logs

Detects defense evasion techniques used by adversaries in endpoint logs including log tampering, timestomping, process injection, and security tool disabling. Use when investigating suspicious endpoint behavior, building detection rules for evasion tactics, or conducting threat hunting for stealthy adversary activity. Activates for requests involving evasion detection, defense evasion analysis, log tampering detection, or MITRE ATT&CK TA0005.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

implementing-microsegmentation-with-guardicore

Implementing microsegmentation using Akamai Guardicore Segmentation to map application dependencies, create granular network policies, visualize east-west traffic flows, and enforce least-privilege communication between workloads across data centers and cloud.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

ramdisk-tmpfs-patterns

暂无简介

★ 13