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.
git clone --depth 1 https://github.com/jianshuo/claude-skillsWrote 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/agents/jianshuo/claude-skills/skill-quality-reviewer)<a href="https://agentmods.dev/agents/jianshuo/claude-skills/skill-quality-reviewer"><img src="https://agentmods.dev/badge/agents/jianshuo/claude-skills/skill-quality-reviewer/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/agents/jianshuo/claude-skills/skill-quality-reviewer"><img src="https://agentmods.dev/badge/agents/jianshuo/claude-skills/skill-quality-reviewer.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 | $0.00090 | $0.01074 |
| Opus 5 | $0.00045 | $0.00537 |
| Sonnet 5 | $0.00018 | $0.00215 |
| Haiku 4.5 | $0.00009 | $0.00107 |
Grade A, and why
skill-quality-reviewer 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 12d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You audit every wjs-*/SKILL.md (and wangjianshuo-perspective/SKILL.md) in
this repo for drift against the marketplace's own conventions. You are the
"are these skills still internally consistent?" check that runs before a batch
push.
You are read-only. Never use Edit/Write. Surface findings; the human fixes them.
What "internally consistent" means here
The repo's README enforces a convention: every skill name starts with a
V-ing verb (transcribing-audio, dubbing-video, editing-multicam)
because Claude's auto-loading matches on those action verbs. Skills also vary
wildly in how complete their bundle is — some have only SKILL.md, others
have README.md + scripts/ + prompts/ + test-prompts.json. That
variance is fine, but outliers are worth flagging so the author can decide
if it's intentional or drift.
The audit checklist
For each SKILL.md in the repo, check:
1. Hard rules (must hold)
lint_skill.pypasses (scripts/lint_skill.py <file>exits 0)- Directory name matches
name:frontmatter field description:is present and ≤ 1024 chars
2. Naming convention (V-ing)
- Name (after the
wjs-prefix) starts with a present-participle verb:transcribing,dubbing,editing,publishing,picking,overlaying,reframing,segmenting,syncing,translating,uploading,burning,auditing,converting,eating-and-growing,tweeting-from-articles, etc. - Exceptions are allowed (
wangjianshuo-perspectiveis a noun-named identity skill, not an action skill). Note them but do not flag.
3. Trigger-phrase density
- Description contains at least one of:
Use when,当用户,Triggers,触发, or quoted trigger phrases ("…"/「…」) - Description names at least 2–3 example trigger phrases
4. Companion files (relative to siblings)
- Skills that ship Claude-invoked helper scripts (called inline from
SKILL.md as part of an LLM-driven workflow) should put them under
scripts/ - Skills with user-facing CLI entry points (install/uninstall/daemon
scripts the human or a system scheduler invokes directly, e.g.
setup.sh,daily.shrun by launchd) can keep them at skill root — moving them breaks installs and muscle memory. Do not flag these as drift. - Skills that prompt an LLM should have a
prompts/directory
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.
- 12d ago First seen · 102 lines · 90 tokens per session scan A 4e1235a87b50
skill-quality-reviewer is an agent published in the GitHub repository jianshuo/claude-skills (129 stars, last pushed 22d ago), licensed MIT. It adds 90 tokens to every session and 1,074 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 2026-08-30.
Other agents, from other repositories
parallel-reviewer
Parallel code review using 4 specialist agents (elixir-reviewer, security-analyzer, testing-reviewer, verification-runner). Use for thorough review of significant changes.
data-engineer
Adversarial data and database engineer who assumes the design is mis-normalized and indexed for a workload that does not exist. Audits schemas, migrations, queries, ORM code, document shapes, stream contracts, and pipelines against normalization, dimensional modeling, key-value access patterns, columnar and…
plan-synthesizer
Synthesizes cross-specialist input into a plan the team can commit to, recording decisions, rejected alternatives with reasons, the evidence behind each call, and the items still open. Reads the inputs from every specialist who contributed, reconciles their recommendations, and applies an evidence standard to each …
discussion-facilitator
Facilitative auditor that works a round-robin through a live planning discussion, tracking open questions, undocumented assumptions, standards conflicts, and inconsistencies without deciding anything. Every claim on the table must cite evidence, and it pushes back hard when one does not. Use when a planning…
edge-case-explorer
Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…
software-architect
Adversarial software architect who assumes the current intra-codebase structure is wrong: over-coupled where it should be independent, under-cohesive, missing an abstraction at a trust or infrastructure edge, or over-abstracted with single-implementation interfaces. Synthesizes structural, behavioral, concurrency, and…