Borrowing it
Nothing to install: this file belongs to dinhnguyenngoc/spec-driven-claude-code. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dinhnguyenngoc/spec-driven-claude-code/main/.claude/commands/verify.mdgit clone --depth 1 https://github.com/dinhnguyenngoc/spec-driven-claude-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/dinhnguyenngoc/spec-driven-claude-code/verify)<a href="https://agentmods.dev/commands/dinhnguyenngoc/spec-driven-claude-code/verify"><img src="https://agentmods.dev/badge/commands/dinhnguyenngoc/spec-driven-claude-code/verify/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/dinhnguyenngoc/spec-driven-claude-code/verify"><img src="https://agentmods.dev/badge/commands/dinhnguyenngoc/spec-driven-claude-code/verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00017 | $0.06308 |
| Opus 5.5 | $0.00007 | $0.02523 |
| Sonnet 5.5 | $0.00003 | $0.01262 |
| Haiku 4.5 | $0.00002 | $0.00631 |
Grade C, and why
verify scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
2. **Let the runner clean its output dir each run.** Playwright clears `outputDir` on start by default → stale artifacts from a previous (e.g. FAILED) run never bleed into the next verdict. If the runner does not auto-cl How it starts
The opening of the file, as written. The whole thing — 267 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/verify — Post-Deploy Real-Environment Verification
"Test what you ship, not a proxy of it."
Purpose
Status: step optional · BLOCKING if run. Strongly recommended before
/deploystages the artifact, especially for brownfield (missing legacy test suite) or releases with infra/config changes. Required when called from the/hotfixorchestrator (Step 4 re-verify on the patched digest).
Workspace Mode: if the session root declares
Mode: workspace→ resolve the target repo perCLAUDE.md§Workspace Mode before anything else; every path, probe, and gate below is relative to the target repo, and the workspace disk-check applies at the gate.
Verifies every user-observable feature works correctly on the exact artifact about to be staged (correct image/build, staging config — the same env that /deploy will use ⇒ within the kit's scope the env always matches, over real network) — before /deploy declares STAGED. The kit stops at staging; promoting to production is a manual step (see deploy.md §The kit's boundary) — config differences between staging↔production belong to the manual checklist DEPLOY_RUNBOOK §8.
/verify closes the fidelity gap that earlier test layers cannot touch: /build (in-memory) and /test (TestContainers + in-process host) both run in a test-environment via in-process transport. A class of bugs only surfaces at (production environment) × (real network) × (real client): CORS, security headers, env-gating middleware, TLS, reverse-proxy headers, container networking, env-var injection, build-time config baking. /verify is the only layer that exercises that intersection.
Scope Clarification
| Command | Responsibility |
|---|---|
/test |
Pre-deploy QA — unit, integration (real deps via TestContainers), contract tests in-process |
/infra |
Setup — Dockerfile, compose, env config |
/verify |
Prove on real artifact — run feature suite against the deployed stack; gate promotion. Step optional · BLOCKING if run (REQUIRED inside /hotfix) |
/deploy |
Promote — if /verify was run, ship only a build with a passing report for that exact digest |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 267 lines · 17 tokens per session scan C 8a7cdf62d687
verify is a command published in the GitHub repository dinhnguyenngoc/spec-driven-claude-code (21 stars, last pushed 2mo ago), licensed MIT. It adds 17 tokens to every session and 6,308 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
fix
Correction automatisee des bugs identifies par la QA Recette.
status
Afficher le statut et la progression des sessions QA Recette, ou générer des rapports (--report).
a11y-audit
Audit Accessibilité WCAG 2.2 AAA.
recette
Tests d'acceptance automatises avec Claude in Chrome.
regression
Voir et gerer le registre de tests de regression QA Recette.
report
Obsolète (retirée en 9.0) : utilisez /qa:status --report.