Dependency Audit logo

Dependency Audit

Community
Mathews-Tom
dependency-audit

Audits direct and transitive dependencies for license compliance, maintenance health, CVEs, abandoned packages, and bloat. Triggers on: "audit dependencies", "license check", "dependency health", "abandoned packages", "unused dependencies", "license compliance", "supply chain", "dependency risk".

Overview

PublisherMathews-Tom
Repositoryarmory
Skill namedependency-audit
Stars
318
Forks
47
Bundled files
5
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Mathews-Tom on GitHub. Read the source before you install it.

Installation

Install the Dependency Audit AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Mathews-Tom/armory.git /tmp/armory
mkdir -p .claude/skills
cp -r /tmp/armory/skills/dependency-audit .claude/skills/dependency-audit
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Dependency Audit in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Dependency Audit on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Dependency Audit is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Dependency Audit

Comprehensive dependency risk assessment: license compatibility analysis, maintenance health scoring, CVE detection, bloat identification, and transitive dependency risk mapping. Produces an actionable report with prioritized remediation steps organized by urgency (security → license → maintenance → bloat).

Reference Files

FileContentsLoad When
references/license-compatibility.mdLicense compatibility matrix, copyleft detection, commercial-safe licensesAlways
references/health-metrics.mdMaintenance health indicators, scoring criteria, abandonment detectionAlways
references/bloat-detection.mdIdentifying unused deps, duplicate functionality, heavy transitive treesBloat analysis requested
references/cve-sources.mdCVE databases, advisory sources, vulnerability severity interpretationSecurity audit requested

Prerequisites

  • Access to the project's dependency files (pyproject.toml, requirements.txt, package.json, Cargo.toml, go.mod)
  • Lock file (for exact versions and transitive dependencies)
  • Project license (to determine compatibility requirements)

Workflow

Phase 1: Parse Dependency Tree

  1. Direct dependencies — Packages explicitly declared in the project.
  2. Transitive dependencies — Dependencies of dependencies. Often 10-50x the direct count.
  3. Version constraints — Pinned (==1.2.3), ranged (>=1.0,<2.0), or floating (*).
  4. Development vs production — Separate dev/test dependencies from production.

Tools:

  • Python: uv pip list, pip-audit, pipdeptree
  • Node.js: npm list --all, npm audit
  • Rust: cargo tree, cargo audit

Phase 2: Audit Licenses

For each dependency:

  1. Identify the license — Check package metadata, LICENSE file, pyproject.toml.

  2. Classify compatibility — Against the project's own license:

    LicenseCommercial OKCopyleftRisk Level
    MIT, BSD, ISC, Apache 2.0YesNoLow
    LGPLWith careWeakMedium
    GPL-2.0, GPL-3.0No (unless GPL project)StrongHigh
    AGPLNo (unless AGPL project)Strong + networkCritical
    UnknownCannot determineUnknownCritical
  3. Flag issues — Copyleft licenses in proprietary projects, unknown licenses, license changes between versions.

Phase 3: Assess Maintenance Health

For each dependency, evaluate maintenance signals:

IndicatorHealthyWarningAbandoned
Last release< 6 months6-18 months> 18 months
Commits (90 days)10+1-90
Open issues response< 2 weeks2-8 weeks> 8 weeks or no response
Bus factor3+ maintainers21
CI statusPassingFlakyFailing or absent

Phase 4: Check Security

  1. Known CVEs — Check against advisory databases:

    • Python: pip-audit, PyPI advisory database
    • Node.js: npm audit, GitHub Advisory Database
    • General: NVD (National Vulnerability Database)
  2. Severity classification — CVSS score interpretation:

    CVSS ScoreSeverityAction
    9.0-10.0CriticalUpgrade immediately
    7.0-8.9HighUpgrade within days
    4.0-6.9MediumUpgrade within weeks
    0.1-3.9LowUpgrade at convenience
  3. Fix availability — Is there a patched version? If not, what's the workaround?

Phase 5: Detect Bloat

  1. Unused dependencies — Dependencies imported nowhere in the codebase.
  2. Duplicate functionality — Multiple packages doing the same thing (2 HTTP clients, 2 JSON parsers).
  3. Heavy transitive trees — Packages that pull in dozens of sub-dependencies for a simple feature.
  4. Size analysis — Large packages used for small functionality.

Phase 6: Report

Produce a prioritized report with action items.

Output Format

text
## Dependency Audit: {Project Name}

### Summary
| Metric | Count |
|--------|-------|
| Direct dependencies | {N} |
| Transitive dependencies | {N} |
| License issues | {N} |
| Maintenance concerns | {N} |
| Security vulnerabilities | {N} |
| Bloat candidates | {N} |

### License Compliance

| Package | Version | License | Compatible | Issue |
|---------|---------|---------|------------|-------|
| {pkg} | {ver} | MIT | Yes | None |
| {pkg} | {ver} | GPL-3.0 | No | Copyleft in proprietary project |
| {pkg} | {ver} | Unknown | Unknown | License not identifiable |

### Maintenance Health

| Package | Last Release | Commits (90d) | Maintainers | Status |
|---------|-------------|---------------|-------------|--------|
| {pkg} | {date} | {N} | {N} | {Healthy/Warning/Abandoned} |

### Security Vulnerabilities

| Package | Version | CVE | Severity | Fix Available | Fixed In |
|---------|---------|-----|----------|---------------|----------|
| {pkg} | {ver} | {CVE-ID} | {severity} | {Yes/No} | {version} |

### Bloat Analysis

| Package | Install Size | Used By | Recommendation |
|---------|-------------|---------|----------------|
| {pkg} | {size} | {usage description} | {Remove/Replace/Keep} |

### Action Items

#### Immediate (Security)
1. Upgrade {pkg} to {version} — fixes {CVE-ID} ({severity})

#### Short-term (License)
1. Review {pkg} GPL usage — may require license change or removal

#### Medium-term (Maintenance)
1. Find alternative to {pkg} — abandoned since {date}

#### Long-term (Bloat)
1. Remove {pkg} — unused in codebase
2. Replace {pkg} with lighter alternative

### Transitive Risk
- {direct-dep} depends on {transitive-dep} which has {issue}

Calibration Rules

  1. Production dependencies first. Dev/test dependencies have lower risk since they don't ship to users. Audit production dependencies with higher scrutiny.
  2. Transitive risk is real. A direct dependency with MIT license may pull in a GPL transitive dependency. Always check the full tree.
  3. Abandoned is not broken. A mature, stable library that hasn't been updated in a year may be perfectly fine. Evaluate based on whether the library is "done" vs "neglected."
  4. Security is non-negotiable. Critical and High CVEs must be addressed immediately. Medium CVEs should be tracked. Low CVEs can wait for the next dependency update cycle.

Error Handling

ProblemResolution
No lock file availableAudit based on declared dependencies. Note that transitive analysis is incomplete without a lock file.
License metadata missingCheck the package's repository for LICENSE file. Note packages where license cannot be determined.
Package registry unavailableWork from cached metadata and local lockfile data.
Too many dependencies to audit manuallyPrioritize: production deps first, then direct deps, then transitive deps with known issues.

When NOT to Audit

Push back if:

  • The project is a prototype that won't ship — defer audit until production decision
  • The user wants dependency updates, not audit — different task (dependabot, renovate)
  • The project has no dependencies (pure standard library) — nothing to audit

Rationalizations

RationalizationReality
"It's a trusted package"Trust is not a security model — trusted packages get compromised (event-stream, ua-parser-js, colors.js)
"Only a minor version bump"Minor versions can introduce vulnerabilities, change behavior, or add transitive dependencies — semver is a promise, not a guarantee
"We don't use the vulnerable function"Transitive dependencies might — and attack surface includes any code loaded into the process
"The CVE is low severity"Low severity in isolation can be critical in your context — a "low" SSRF in an internal service with cloud metadata access is critical
"We'll update when there's a known exploit"Known exploits mean you're already behind — patch within SLA, not after breach
"Too many dependencies to audit"That's the problem, not an excuse — high dependency count IS a risk finding

Red Flags

  • Auditing only direct dependencies while ignoring transitive dependency tree
  • Dismissing CVEs without checking if the vulnerable code path is reachable
  • No license compatibility check — GPL in a proprietary codebase is a legal finding
  • Accepting "no known vulnerabilities" from a single scanner without cross-referencing
  • Ignoring dependency age — unmaintained packages with no updates in 2+ years are a risk
  • Skipping lockfile analysis (pinned vs. floating versions)

Verification

  • Both direct and transitive dependencies scanned
  • Vulnerability scanner output captured: npm audit / pip-audit / cargo audit
  • Each CVE finding includes: severity, affected version range, upgrade path, reachability assessment
  • License compatibility verified against project license
  • Dependency age and maintenance status checked for top-level deps
  • Lockfile present and version pinning verified — no floating ranges in production

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Dependency Audit AI skill do?

Audits direct and transitive dependencies for license compliance, maintenance health, CVEs, abandoned packages, and bloat. Triggers on: "audit dependencies", "license check", "dependency health", "abandoned packages", "unused dependencies", "license compliance", "supply chain", "dependency risk".

Why use Dependency Audit on TypingMind?

Because you install it once and use it with any model. Dependency Audit is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Dependency Audit in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Mathews-Tom/armory/tree/main/skills/dependency-audit. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Dependency Audit?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Dependency Audit?

As many as you like. As long as a model supports skills, you can use Dependency Audit with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Dependency Audit AI skill free?

Yes. It is published on GitHub by Mathews-Tom under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇