skill-vetter

Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

The canonical home for this skill is skill-vetter in netease-youdao/LobsterAI

Quality

Content

85%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with concrete commands, checklists, and templates, plus a well-sequenced vetting workflow. Its main weakness is conciseness: decorative ASCII framing and a redundant summary section add tokens without adding information.

Suggestions

Drop the box-drawing borders in the code blocks or replace them with plain fenced lists; the visual framing costs tokens that the checklist content already conveys.

Merge the 'Remember' section into the relevant protocol steps (e.g., fold 'ask your human for high-risk decisions' into Step 4) to eliminate restated guidance.

Dimension	Reasoning	Score
Conciseness	The body is mostly token-efficient checklists and tables, but the decorative box-drawing borders and the 'Remember' section (which restates earlier guidance) add overhead that could be tightened, matching the score-2 anchor rather than the lean score-3 example.	2 / 3
Actionability	It provides copy-paste-ready curl commands, a concrete red-flag checklist, a risk-classification table, and a full output-report template, matching the fully-executable score-3 anchor.	3 / 3
Workflow Clarity	The vetting protocol is a clearly sequenced four-step process with a mandatory code-review checkpoint, checklists at each step, and a risk-level-to-verdict decision mapping, matching the score-3 anchor with explicit checkpoints.	3 / 3
Progressive Disclosure	The skill is a single self-contained file (no references/scripts/assets bundle exists) with well-organized sections and no nested references, so per the simple-skills note it qualifies for the score-3 anchor on clear organization.	3 / 3
	Total	11 / 12 Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it concisely states purpose, gives an explicit use-trigger tied to concrete sources, and occupies a distinct security-vetting niche. The only weakness is that its named capabilities are somewhat abstract categories rather than a granular action list.

Suggestions

Replace abstract categories ('suspicious patterns') with one or two concrete checks (e.g., 'flags credential access, data exfiltration, and obfuscated code') to lift specificity toward a score of 3.

Dimension	Reasoning	Score
Specificity	The description names the vetting domain and several actions ('Checks for red flags, permission scope, and suspicious patterns'), but these are broad categories rather than a comprehensive list of concrete actions, matching the score-2 anchor.	2 / 3
Completeness	It explicitly answers both what it does ('Checks for red flags, permission scope, and suspicious patterns') and when to use it ('Use before installing any skill...'), with an explicit trigger clause, matching the score-3 anchor.	3 / 3
Trigger Term Quality	It uses natural trigger phrases a user would say ('installing any skill from ClawdHub, GitHub, or other sources') with good coverage of concrete source terms, matching the score-3 anchor and the 'Use when...' good-example pattern.	3 / 3
Distinctiveness Conflict Risk	'Security-first skill vetting' carves a clear niche with distinct install-from-external-source triggers unlikely to conflict with other skills, matching the score-3 anchor.	3 / 3
	Total	11 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: UnicomAI/wanwu
Path: configs/microservice/bff-service/configs/agent-skills/security/skill-vetter/SKILL.md
Commit: f618458

Reviewed: about 12 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.