Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/hmj1026/dhpknpx agentmods add commands/hmj1026/dhpk/verifyWrote 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/hmj1026/dhpk/verify)<a href="https://agentmods.dev/commands/hmj1026/dhpk/verify"><img src="https://agentmods.dev/badge/commands/hmj1026/dhpk/verify.svg" alt="Measured on agentmods" 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 | $0.00027 | $0.01211 |
| Opus 5 | $0.00014 | $0.00606 |
| Sonnet 5 | $0.00005 | $0.00242 |
| Haiku 4.5 | $0.00003 | $0.00121 |
Grade A, and why
verify scanned grade A with 0 findings 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 yesterday.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/verify — Verification loop
Run one read-only verification loop against the current repository. Resolve the mode first, use the installed runner when present, and report every applicable stage as PASS, FAIL, or SKIP.
Context
- Branch: !
git branch --show-current - Changes: !
git diff --stat HEAD
Step 1 — normalize the request
- No mode or
full→ run lint, typecheck, unit, integration, and e2e stages. fast→ run lint and unit stages only.--integration <path>and--e2e <path>opt into one explicit test file in the corresponding full-mode stage.- Accept one mode token,
fastorfull; report usage and stop on any other mode token.
Completion criterion: the mode is fast or full, and each optional test path
is either absent or attached to its matching stage.
Step 2 — use the installed runner
Use Glob to check for .claude/scripts/verify-runner.js at the project root.
When found, run:
node .claude/scripts/verify-runner.js $ARGUMENTS
The runner owns package-manager detection, stage order, graceful skips, logs, and the final summary. Treat its output as authoritative:
- An overall line containing
PASS→ verification passed. - An overall line containing
FAIL→ verification failed; report the failed stage(s) and their log paths. runner crashed, a missing overall line, or an execution error → report a runner failure.
Runner failure remains terminal. Select the fallback only when the runner file is absent.
Completion criterion: the runner output contains one overall verdict and every executed or skipped stage is represented in that summary.
Step 3 — fallback when the runner is absent
Detect the first matching project manifest and run the commands in stage order.
For Node.js, detect the package manager from its lockfile and read
package.json before choosing scripts.
| Manifest | Ecosystem | Fast | Full additions |
|---|---|---|---|
package.json |
Node.js | {pm} lint; {pm} test:js, then test:unit, then test |
{pm} typecheck or local npx --no-install tsc --noEmit; explicit test:integration and test:e2e paths |
pyproject.toml |
Python | ruff check .; pytest |
mypy . |
Cargo.toml |
Rust | cargo clippy; cargo test |
Rust compilation/typechecking is part of these commands |
go.mod |
Go | golangci-lint run; go test ./... |
go vet ./... |
build.gradle or pom.xml |
Java | project lint task when present; project test task | build/typecheck is implicit in the build or verification task |
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.
- yesterday First seen · 118 lines · 27 tokens per session scan A 04686ded98ec
verify is a command published in the GitHub repository hmj1026/dhpk (2 stars, last pushed today), licensed MIT. It adds 27 tokens to every session and 1,211 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other commands, from other repositories
cloudflare-workers:test-setup
Interactive Vitest setup wizard for Cloudflare Workers testing. Configures @cloudflare/vitest-pool-workers, mocks bindings, and creates example tests.
sdlc-test
Generate and run multi-tiered enterprise test suites (Unit, Integration, E2E, Contract, Load/k6, Chaos, and Fuzz testing).
generate-tests
Generate unit, integration, and e2e test suites from scaffolded project with framework-appropriate coverage.
flutter-test
Run Flutter/Dart tests, report failures, and incrementally fix test issues. Covers unit, widget, golden, and integration tests.
add-tests
Generate comprehensive tests for a component or API endpoint.
cloudflare-durable-objects:setup
Interactive Durable Objects project initialization. Scaffolds new DO project with proper bindings, migrations, and boilerplate code.