Borrowing it
Nothing to install: this file belongs to hookdeck/hookdeck-cli. 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/hookdeck/hookdeck-cli/main/.agents/skills/hookdeck-cli-review/SKILL.mdgit clone --depth 1 https://github.com/hookdeck/hookdeck-cliWrote 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/skills/hookdeck/hookdeck-cli/hookdeck-cli-review)<a href="https://agentmods.dev/skills/hookdeck/hookdeck-cli/hookdeck-cli-review"><img src="https://agentmods.dev/badge/skills/hookdeck/hookdeck-cli/hookdeck-cli-review/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/skills/hookdeck/hookdeck-cli/hookdeck-cli-review"><img src="https://agentmods.dev/badge/skills/hookdeck/hookdeck-cli/hookdeck-cli-review.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 | $0.00085 | $0.04734 |
| Opus 5.5 | $0.00034 | $0.01894 |
| Sonnet 5.5 | $0.00017 | $0.00947 |
| Haiku 4.5 | $0.00009 | $0.00473 |
Grade A, and why
hookdeck-cli-review 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 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.
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 — 416 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reviewing changes to the Hookdeck CLI
AGENTS.md is the reference for building, testing and linting this repository,
and for the rules it already states. Read it first. This file is the review
delta: what to look at in a diff, and the traps that a green pull request
hides. Where a rule already lives in AGENTS.md, this points at it rather than
restating it.
Acceptance runs on the pull request, but nothing enforces it
.github/workflows/test-acceptance.yml triggers on every pull_request
targeting main or next, with no path filter. So the acceptance suite — the
three matrix slices plus the separate telemetry job — does run on the pull
request in front of you. Assuming it did not is the wrong call.
What it is not is required. Branch protection on main requires exactly:
unit-test build-linux build-mac build-windows
Acceptance is not in that list. A pull request with a red or skipped acceptance run is still mergeable, and the merge button will not tell you.
That list is the one claim here with no source in the tree — it lives in GitHub's branch-protection settings, so it can change without a commit and without this file noticing. Re-read it rather than trusting it:
gh api repos/hookdeck/hookdeck-cli/branches/main/protection \
--jq '.required_status_checks.contexts'
So read gh pr checks and flag a red or skipped acceptance run, because
nothing else in the process will. That is the single most useful thing this
review does that the required checks do not.
A green acceptance job is not a run
A tick is not coverage either. An acceptance test that needs a credential skips
itself when the job's env: block does not supply one, and a job whose tests
all skipped still reports green. gh pr checks cannot show the difference.
acceptance-telemetry is the live example — see CLITelemetry JSON tags are a
wire contract below.
Three gates sit between an assertion existing and a red check, and all three
have to hold: the file's build tag has to be in a slice list or the telemetry
job's -tags; the test's own guards (testing.Short(), a t.Skip on a missing
environment variable) have to pass; and the job's env: block has to supply
what those guards need.
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 Changed · +12 lines 503f3ed75346
- 1mo ago First seen · 404 lines · 85 tokens per session scan A 9cf982214d02
hookdeck-cli-review is a skill published in the GitHub repository hookdeck/hookdeck-cli (366 stars, last pushed yesterday), licensed Apache-2.0. It adds 85 tokens to every session and 4,734 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-25.
Other skills, from other repositories
acceptance
Test a PR review end to end and prove it honoured every guarantee in INVARIANTS.md. Use when asked to test/validate a review of a pull request, to check a finished run, or to run the 3-provider x 2-runtime acceptance matrix. Drives pr-review verify and scripts/acceptance.mjs — never hand-checks the guarantees.
e2e-testing
Guide for running end-to-end tests of the Qwen Code CLI, including headless mode, MCP server testing, and API traffic inspection. Use this skill whenever you need to verify CLI behavior with real model calls, reproduce user-reported bugs end-to-end, test MCP tool integrations, or inspect raw API request/response…
test-conversion-reviewer
Reviewer for converting unit tests to browser tests for Project Bedrock. Invoke this when the user needs to review their work while removing complex Browser dependencies from tests.
verify
Exercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.
webapp-testing
Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
review-work
Post-implementation gate review: run manual QA on the real surface yourself, then launch ONE gate reviewer (never a panel) to audit goal, constraints, code quality, security, missed context, and QA evidence. Use before a PR handoff or when the user explicitly asks to review completed work.