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 pandysp/claude-plugins --skill quality-reviewgit clone --depth 1 https://github.com/pandysp/claude-pluginsWrote 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/pandysp/claude-plugins/quality-review)<a href="https://agentmods.dev/skills/pandysp/claude-plugins/quality-review"><img src="https://agentmods.dev/badge/skills/pandysp/claude-plugins/quality-review.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 | $0.00095 | $0.02565 |
| Opus 5 | $0.00048 | $0.01282 |
| Sonnet 5 | $0.00019 | $0.00513 |
| Haiku 4.5 | $0.00010 | $0.00257 |
Grade A, and why
quality-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 7d 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/quality-review — multi-lens quality audit
Usage: /quality-review [level] [--fix] [--domain docs|code] <target>
- level:
low|medium|high|xhigh|max. If no level is given, use the session's current effort level (what /effort reports); if it cannot be determined, usemedium. Low and medium run inline in this conversation, tuned for fewer, high-confidence findings. High and above launch the audit workflow. Xhigh and max add larger candidate budgets and a gap sweep: one extra finder re-reads the files for defects the per-lens pass missed, which only pays off once that pass is already thorough. - --fix: after the review, apply the findings to the working tree. Skip findings that are wrong or not worth fixing, and run the relevant checks after (for docs, the writing-voice self-test if that skill is available).
- --domain: force the lens sheet. Without it, infer from the target: prose and markdown files are
docs, source files arecode. If the target mixes both or fits neither, ask. - target: files, directories, or a description of what to audit. Required; if missing, ask what to audit rather than guessing.
The 13 adjectives
| Adjective | Meaning |
|---|---|
| unity | One idea runs the whole thing and every part serves it. Parts agree and don't contradict. Reads like one person wrote it: same voice, same terms throughout. |
| vividness | The reader sees it, not just understands it. One well-chosen concrete detail. Things happen instead of sitting there. |
| authority | The text shows command. Credentials don't count. Claims come with visible backing. Confidence matches the evidence. Takes a position you could hold it to. |
| economy | Every element earns its place. Cut it and something is lost. Good parts do several jobs at once. Nothing added for show. |
| sensitivity | Register and tone fit this audience and this occasion. Doesn't lecture or presume. Knows what to leave unsaid. This is the social fit. Whether they can parse it is clarity's job. |
| clarity | Nothing between the reader and the point. Only one way to read it, and you get it first pass. Actors are subjects and actions are verbs. Uses words the reader already knows. |
| emphasis | One element dominates and the rest step back. The key point goes where attention is highest. Stands out by contrast. Emphasize only a few things. More and none of them stand out. |
| flow | Each unit starts from what the reader knows and ends on the new thing. Seamless joins. The reader keeps going because it's easy, not because they're waiting for an answer. |
| suspense | Raises a question the reader wants answered. Reveals in doses. Pays off what it set up. |
| brilliance | Surprising and right at the same time. One without the other is worthless. An unexpected connection that holds. Comes on top of the basics, never instead of them. Sits in one spot, not spread across the whole. |
| precision | The right word, not the almost-right word. Vague is a defect here. Ambiguous is clarity's. The narrowest term that is still true. Says all, most, or some and means it. |
| proportion | Space tracks importance. Part-to-whole ratios are right. No minor point gets a page and no major point gets a sentence. |
| depth | More under the surface than on it. Rereading finds another layer and the first reading was already right. Goes deep on one thing instead of touching many. Push on any claim and there's more behind it. Not a front. |
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 90 lines · 95 tokens per session scan A 53583dbce5ed
quality-review is a skill published in the GitHub repository pandysp/claude-plugins (5 stars, last pushed today), licensed MIT. It adds 95 tokens to every session and 2,565 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
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"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
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…