Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add sharpTrick/parley --skill critical-reviewgit clone --depth 1 https://github.com/sharpTrick/parleyWrote 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/sharptrick/parley/critical-review)<a href="https://agentmods.dev/skills/sharptrick/parley/critical-review"><img src="https://agentmods.dev/badge/skills/sharptrick/parley/critical-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/sharptrick/parley/critical-review"><img src="https://agentmods.dev/badge/skills/sharptrick/parley/critical-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 | $0.00079 | $0.01653 |
| Opus 5.5 | $0.00032 | $0.00661 |
| Sonnet 5.5 | $0.00016 | $0.00331 |
| Haiku 4.5 | $0.00008 | $0.00165 |
Grade A, and why
critical-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 5d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Critical review (Careening)
The operator's guide. The process is docs/REVIEW_PROTOCOL.md; the reviewer is
.claude/agents/critic-package.md; the runner is .claude/workflows/careening-review.js. Read the
protocol before your first round — the anti-patterns in it are all things that have already gone
wrong somewhere.
Run a round
Workflow({ scriptPath: "<repo>/.claude/workflows/careening-review.js",
args: { round: N, quiesced: [...], changed: [...], wakeAll: false,
worktrees: "...", base: "<sha>" } })
round— the round number, for labels and the record.quiesced— pass back the previous round'snextQuiescedverbatim.changed—git diff --name-only <last-round-sha>..HEAD. This drives wake-up.wakeAll— required for any round that may declare convergence.worktrees— required. The base directorynode scripts/careening-worktrees.mjs setup <sha>provisioned under, which it prints on its last line. There is no default: the runner refuses the round rather than send fifteen critics somewhere it guessed.base— the round's base sha. Each critic checks its own worktree against it before reading.
Round 1: still needs worktrees and base. Every critic runs, nothing is asleep yet.
Read the result
| field | what to do with it |
|---|---|
converged |
true only from a wake-all round with zero CONFIRMED. Two consecutive → done. |
blockingClean |
The shadow signal. Record it; do not act on it. |
counts |
confirmed / blocking / plausible / total. |
confirmed |
The work list. Fix or explicitly justify every one. |
plausible |
Adjudicate: fix, or write down why not. Doesn't block convergence. |
nextQuiesced |
Pass straight back in as the next round's quiesced. |
byLens |
Per-lens yield — the experiment's primary measurement. |
byTarget |
Per-critic counts, plus checked for the ones that found nothing. |
Remediating a round
Past the first few findings, do not fix them inline. Fan out one remediation agent per package, hand it that package's findings, and keep the orchestrator's context for adjudication and the commit. A round can produce a hundred findings; reading each one's code into a single context is what forces a mid-run compaction, and an orchestrator that has forgotten round 3 cannot notice that round 9 is re-finding it.
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.
- 5d ago First seen · 117 lines · 79 tokens per session scan A bdc0c90546c7
critical-review is a skill published in the GitHub repository sharpTrick/parley (0 stars, last pushed 5d ago), licensed MIT. It adds 79 tokens to every session and 1,653 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 2026-10-26.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…