More screenshots
Works with the AI Agents you already use
dependency supply chain red flag checker
Offline lockfile/manifest review: install scripts, typosquats vs bundled popular lists, git/URL deps, unpinned ranges.
$15
dependency supply chain red flag checker
Example session with this skill installed
Before I npm install this repo — any typosquats or postinstall scripts I should worry about?
- Read your context and instructions
- Compiled the dependency supply chain
Answer: 2 CRITICAL, 2 HIGH, 4 MEDIUM on the sample app manifests.
- CRITICAL (npm):
lodahs@4.17.21is Damerau-Levenshtein distance 1 from popular packagelodash - CRITICAL (pypi):
requets==2.31.0is distance 1 from popularrequests - HIGH:
package.jsondefines apostinstalllifecycle script (node ./scripts/fetch-assets.js) - HIGH:
left-paddepends viagit+https://github.com/evil.example/left-pad.git(bypasses the registry) - MEDIUM: floating versions
express=latest,react=^18.2.0; unpinnedFlaskandbeautifulsoup4>=4.12
Popular lists loaded from the bundle
npm=3000,
pypi=3000, cargo=103, go=88 (sources cited in references/popular/README.md).
Deliverables: severity-sorted red flags; optional deps.md + JSON. Never installs packages. Never hits the network.
Next steps: verify typosquat hits are intentional before removing; pin versions; drop git/URL deps or vendor them deliberately; review install scripts. A typosquat hit is a signal, not proof of malware.
(Real run on bundled examples/project/.)
Connects securely to your tools. The creator never sees your data.
About this skill
Red-flag your dependencies before you install them.
Reviews package.json / package-lock.json, requirements.txt, pyproject.toml, Cargo.toml and go.mod for lifecycle/install scripts, typosquat distance ≤1 against a bundled popular-package list (3,000 npm + 3,000 PyPI + curated crates/Go), git/URL/tarball deps, and floating versions. Never installs packages. Never hits the network.
What's included
SKILL.md, scripts/depflags.py, references/popular/* with cited sources, sample project.
- Typosquat ≠ confirmed malware — verify intent before removing.
- Popular lists are snapshots (see references/popular/README.md); refresh periodically.
- yarn.lock / pnpm-lock deep parsing not implemented (convert or use package-lock).
- Does not query advisories/CVEs (no network).
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 any payment.
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 29 days ago
Requires
Permissions
File Scopes
Fully offline. Scripts never import networking or subprocess modules and never modify system or cloud configuration.