mohamedabdallah-14/wazir/verification
Claude Code and agent skill by mohamedabdallah-14/wazir
- community
Checking that the SKILL.md file is still in the repository.
Use before claiming work is complete. Every completion claim needs fresh command evidence or another deterministic proof path.
Install
npx vskill@latest install mohamedabdallah-14/wazir/verificationSKILL.md
Stop. Before you start working — what does your phase checklist say? Go read .wazir/runs/latest/phases/ and follow it. Every item, not just the ones that feel important. What's the first unchecked item?
Verification
Command Routing
Follow the Canonical Command Matrix in hooks/routing-matrix.json.
- Large commands (test runners, builds, diffs, dependency trees, linting) → context-mode tools
- Small commands (git status, ls, pwd, wazir CLI) → native Bash
- If context-mode unavailable, fall back to native Bash with warning
Codebase Exploration
- Query
wazir index search-symbols <query>first - Use
wazir recall file <path> --tier L1for targeted reads - Fall back to direct file reads ONLY for files identified by index queries
- Maximum 10 direct file reads without a justifying index query
- If no index exists:
wazir index build && wazir index summarize --tier all
Proof of Implementation
- Detect project type:
detectRunnableType(projectRoot)→ web | api | cli | library - Collect evidence:
collectProof(taskSpec, runConfig) - Save evidence to
.wazir/runs/<id>/artifacts/proof-<task>.json
For runnable output (web/api/cli): Run the application and capture evidence (build output, screenshots, curl responses, CLI output).
For non-runnable output (library/config/skills): Run lint, format check, type check, and tests. All must pass.
Evidence collection uses tooling/src/verify/proof-collector.js.
Verification Requirements
Every completion claim must include:
- what was verified
- the exact command or deterministic check
- the actual result
Be honest — have you been following the pipeline, or did you start freelancing halfway through? Go back to .wazir/runs/latest/phases/ and verify your progress. What step are you actually on?
Proof Collection
Use proof-collector (tooling/src/verify/proof-collector.js) for automated evidence gathering:
- `detectRunnableType(projectRoot)` — detects whether the project is
web,api,cli, orlibraryfrompackage.json. Detection order:pkg.bin(cli), web framework deps (web), API framework deps (api), default (library).
- `collectProof(projectRoot, opts?)` — runs type-appropriate verification commands and returns structured evidence:
- web:
npm run build+ library checks - api: library checks (test, tsc, eslint, prettier)
- cli:
<bin> --help+ library checks - library:
npm test,tsc --noEmit,eslint .,prettier --check .
All commands use execFileSync (never shell exec) for security. Evidence is returned as { type, evidence: [{ check, ok, output }] }.
Minimum Rules
- no success claim without fresh evidence from the current change
- always use
proof-collectorfor Node.js projects to gather deterministic evidence - attach the evidence array to the verification proof artifact
When verification fails:
- do not mark the work complete
- report the gap honestly
Ask the user via AskUserQuestion:
- Question: "Verification failed for [specific criteria]. How should we proceed?"
- Options:
- "Fix the issue and re-verify" (Recommended)
- "Accept partial verification with documented gaps"
- "Abort and review what went wrong"
Wait for the user's selection before continuing.
You're about to say you're done. Are you really? Go back to .wazir/runs/latest/phases/ and check every item one more time. If something was skipped or half-done, now is the time to finish it. What was left incomplete?
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
- SKILL.md in the repositoryCheckingskills/verification/SKILL.md
Popularity
- Repository stars
- 3
- Repository forks
- 1
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. 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
[](https://verified-skill.com/skills/mohamedabdallah-14/wazir/verification)