More screenshots
Works with the AI Agents you already use
Dependency & License Risk Auditor
Audits packages for CVEs, abandonment, and license risk before any agent adds them — blocks GPL creep and supply-chain surprises.
$49
Dependency & License Risk Auditor
Example session with this skill installed
I'm about to add lodash@4.17.21 to our commercial Node project (proprietary license).
Run the full Dependency & License Risk Auditor before I install it.
- Read your context and instructions
- Compiled the dependency & license
Dependency Risk Audit Report
Package: npm / lodash @ 4.17.21
Project license target: proprietary
Audit timestamp: 2026-09-06T09:24:25Z
Auditor version: 1.0.0
Overall Recommendation
BLOCK — do not add this package
Risk level: critical
1. Known Vulnerabilities (OSV.dev)
- Count: 3
- Highest severity: HIGH
- IDs: GHSA-f23m-r3pf-42rh, GHSA-r5fr-rjxr-66jc, GHSA-xxjr-mmjv-4gpg
| ID | Severity | Summary |
|---|---|---|
| GHSA-r5fr-rjxr-66jc | HIGH | lodash vulnerable to Code Injection via _.template imports key names |
| GHSA-f23m-r3pf-42rh | MEDIUM | Prototype Pollution via array path bypass in _.unset and _.omit |
| GHSA-xxjr-mmjv-4gpg | MEDIUM | Prototype Pollution in _.unset and _.omit |
2. Maintenance Status
- Last publish: 2023 days ago (> 24 months)
- Weekly downloads: 173M (still popular, but this specific version is stale)
- Latest available: 4.18.1
- Verdict: Strong abandonment signal on the requested version.
3. License Assessment
- Declared: MIT
- Normalized: MIT
- Compatibility: compatible
- Risk: low
- Rationale: Permissive license. Safe for commercial use with attribution.
Decision
Blocked. High-severity CVE + very old version.
Safer alternatives that pass the same audit:
lodash@4.18.1(or latest) — re-run the auditor- Native ES alternatives or
es-toolkitfor tree-shakable utilities - Pin only the specific functions you need via a maintained fork
Next steps
- Do not run
npm install lodash@4.17.21 - Re-audit any candidate replacement with the same skill
- Attach this report to the PR / change request for audit trail
Connects securely to your tools. The creator never sees your data.
About this skill
Stop the expensive “boring” failures before they hit production.
Before an agent runs npm install, pip install, poetry add, or any dependency change, this skill forces a structured risk audit:
- Known CVEs via OSV.dev (exact version)
- Abandoned / unmaintained signals (last publish, downloads, missing versions)
- License compatibility (strong copyleft vs proprietary, weak copyleft, unknown licenses)
Returns a clear gate: Block / Warn / Flag / Allow + structured JSON + human report.
Why buyers need it
- One high-severity CVE or accidental GPL in a commercial codebase costs far more than this skill.
- Natural sibling to Schema Guardrail — run this before the install, Schema Guardrail after.
- Works with any SKILL.md-compatible agent (Claude Code, Cursor, Codex, Gemini CLI, etc.).
- Zero API keys required for the core checks. Only needs Python 3.10+ and requests.
Perfect for agencies shipping coding agents to commercial clients who cannot afford license contamination or silent abandoned packages.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Currently provides full maintenance + license metadata only for npm and PyPI. Other ecosystems (Go, crates.io, Maven, RubyGems, NuGet) receive solid CVE coverage via OSV but limited registry signals.
Abandonment detection is intentionally conservative (>24 months since last publish = High). Some intentionally “finished” utility packages may be flagged; human override is always available.
Does not automatically traverse or audit the full transitive dependency tree unless a lockfile is supplied and the agent explicitly runs batch checks.
Requires outbound HTTPS access to public endpoints (OSV.dev + package registries). Offline / air-gapped environments are not supported in v1.
License assessment uses declared SPDX-style identifiers and a fixed commercial policy matrix. Unusual or dual-licensed packages may need manual legal review.
Network failures or temporary registry outages fail closed (treated as elevated risk) rather than silently allowing the package.
The included script needs Python 3.10+ and the requests library. No heavier dependencies are required.
These constraints are accurate for the current 1.0.0 release and help set correct buyer expectations.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago
- Accounts
Osv
- Software
- Skills Over MCP
Python
Permissions
Allowed Hosts