CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dependency-audit

Audit project dependencies, frameworks, languages, and dev tools for known vulnerabilities, CVEs, and security anti-patterns. Use when the user mentions 'dependency audit,' 'npm audit,' 'CVE,' 'vulnerable packages,' 'supply chain security,' 'outdated dependencies,' 'known vulnerabilities,' 'security advisory,' 'package security,' 'framework vulnerability,' 'is this package safe,' 'pnpm audit,' 'yarn audit,' 'bundle audit,' 'patch this CVE,' 'audit fix broke,' 'lockfile diff,' 'dependency confusion,' 'registry redirection,' 'MCP server pinning,' 'release pipeline,' or needs to check whether their stack, CI, or release pipeline has known security issues.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

81%Weight 40%Scale 1-5

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

An expert-level, highly actionable skill body with an exceptional validation discipline: exact commands, greps, severity calibration, rejection rules, and a validated JSON findings contract. Its main weakness is token weight — long inline catalogs of framework anti-patterns Claude largely already knows, plus inline time-sensitive version details that will rot.

Suggestions

Move the framework-specific anti-pattern catalogs (Next.js/Django/Rails/Express/Spring/Laravel/WordPress) into a reference file (e.g. references/framework-patterns.md), keeping only a pointer plus the genuinely non-obvious items (server-only import checks, in-memory state ≠ rate limit) inline.

Relocate time-sensitive version and date details (CVE-2025-29927 version cutoffs, Go ≥ 1.24 MPTCP, libvips ≥ 8.13, Ruby json 2→3 breakage, October 2026 field notes) into the dated references they already point to, or a clearly marked old-patterns section, so they can be updated without degrading the body.

Trim common-knowledge checklist entries (typosquatting and malicious-package definitions, "DEBUG=True in production," "Default admin username, weak passwords") down to the non-obvious field-tested gotchas that Claude would not already know.

DimensionReasoningScore

Conciseness

The body is dense with little prose padding, but the ~60-line framework anti-pattern catalogs include common-knowledge items Claude already knows ("DEBUG=True in production," "Default admin username, weak passwords," typosquatting definitions), and time-sensitive version/date details are inline rather than in a dated or old-patterns section (CVE-2025-29927, "Go ≥ 1.24 listens with MPTCP," libvips ≥ 8.13, "October 2026" field notes). This is anchor 3 — mostly efficient but with unnecessary explanation and tightening opportunities — above 2 (no tutorial-style filler) but below 4 (more than minor trimming needed).

3 / 5

Actionability

Fully executable throughout: copy-paste-ready audit commands per ecosystem (npm/pnpm/yarn/pip/bundle/govulncheck/composer/dotnet/trivy), exact grep regexes for source redirection and unpinned MCP servers, a dry-run-before---force fix sequence with reachability triage steps, a complete findings.json report template, and a validation script invocation ("node references/validate-findings.mjs findings.json"). Matches anchor 5.

5 / 5

Workflow Clarity

Steps 0–5b are clearly sequenced with explicit validation checkpoints and feedback loops: "run node references/validate-findings.mjs findings.json until it exits 0," the verify-after-fixing checklist (advisory gone on re-audit, no downgrades in the lockfile diff, clean frozen install, full tests, preview smoke test), and dry-run gates before lockfile-rewriting commands. The destructive/batch cap does not apply because validation is explicit and thorough; matches anchor 5.

5 / 5

Progressive Disclosure

All ten referenced files exist, are one level deep, and are well signaled with section pointers and content descriptions ("Details: references/fixing-safely.md §10," "Greps: references/source-redirection.md §2"), with sibling cross-links among reference files rather than nested chains. However the body itself runs ~410 lines and carries large inline detail catalogs (framework-specific patterns, the 8-row CI handoff table, the full report template) beyond overview scale — anchor 4 with minor organization gaps, not 5.

4 / 5

Total

17

/

20

Passed

Description

91%Weight 40%Scale 1-5

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

A strong description: explicit what-and-when structure, third-person voice, and an unusually comprehensive set of natural trigger phrases including tool-specific command names and user-failure phrasing. The only weaknesses are a slightly monolithic action verb ("audit") and a couple of trigger terms that overlap with closely related skills in the same family.

DimensionReasoningScore

Specificity

"Audit project dependencies, frameworks, languages, and dev tools for known vulnerabilities, CVEs, and security anti-patterns" and "check whether their stack, CI, or release pipeline has known security issues" name concrete actions with broad object coverage, but only about two distinct verbs versus the anchor-5 example's list of multiple distinct actions. Fits anchor 4 (several specific elements, minor coverage gaps) rather than 5; not 3 because the domain and actions are explicitly concrete.

4 / 5

Completeness

It clearly answers "what" ("Audit project dependencies, frameworks, languages, and dev tools for known vulnerabilities, CVEs, and security anti-patterns") and explicitly answers "when" ("Use when the user mentions ... or needs to check whether their stack, CI, or release pipeline has known security issues") with concrete trigger phrases. This is a direct match to the anchor-5 example pattern.

5 / 5

Trigger Term Quality

The description lists 21+ natural trigger phrases including tool-specific variants ("npm audit," "pnpm audit," "yarn audit," "bundle audit"), a natural user question ("is this package safe"), and failure phrasing ("audit fix broke") — comprehensive coverage of natural terms and synonyms, matching the anchor-5 example's breadth. Not 4 because no commonly-said variant in this domain is obviously missing.

5 / 5

Distinctiveness Conflict Risk

The dependency/stack security niche is clear with distinct triggers (lockfile diff, dependency confusion, registry redirection), giving minimal conflict with unrelated skills. However "release pipeline" and "MCP server pinning" are triggers this skill defers to sibling skills (cicd-security, agent-supply-chain-audit), creating minor overlap with closely related skills — anchor 4 rather than 5.

4 / 5

Total

18

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository

briiirussell/cybersecurity-skills

Path

skills/dependency-audit/SKILL.md

Reviewed

Table of Contents

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.