Aidevshield NPM audit for AI coding tool workflows

Name: Aidevshield NPM audit for AI coding tool workflows
Availability: InStock
Author: GrimLabs

by GrimLabs·Mar 6, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●●BangerSolve My ProblemBig Brain

Scans Cursor/Cline/GitHub Actions for prompt injection and supply chain attacks—aidevshield catches real exploits.

Strengths

•Targets genuine 2025 attacks (Clinejection, Shai-Hulud, tag repointing)—not theoretical—with working real-world proof
•One command, zero external dependencies beyond js-yaml, fully offline—no setup friction or telemetry risk
•Unified scanning across 4 domains (workflows, AI configs, npm, prompt injection) where competitors only cover 1-2 each

Weaknesses

•Early GitHub repo with minimal stars (1)—zero proof of adoption or community validation yet
•Rule coverage for npm/postinstall attacks less detailed than Socket.dev; unclear how rule updates ship

Post Description

I built aidevshield, a free open-source security scanner that catches vulnerabilities in AI coding tool configurations things like Cursor, Copilot, and Cline setups.

The problem: AI coding assistants are becoming attack surfaces. In the past year we've seen real exploits:

Clinejection (Dec 2025): Prompt injection via GitHub issue titles → AI triage bot executes shell commands → malicious npm package published to 5M users tj-actions/changed-files (Mar 2025): Tag repointing attack hit 23,000 repos via pull_request_target Shai-Hulud (2025): First self-propagating npm worm using postinstall hooks Rules File Backdoor (Feb 2025): Hidden Unicode characters in .cursorrules for invisible prompt injection Existing tools cover parts of this — Zizmor and actionlint for GitHub Actions, Socket.dev for npm — but nothing scans across all four domains: workflows + AI configs + npm packages + prompt injection.

aidevshield does. One command, zero dependencies beyond js-yaml, fully offline:

npx aidevshield scan . What it detects:

Dangerous GitHub Actions patterns (wildcard users, untrusted event field interpolation, unpinned third-party actions) pull_request_target with untrusted checkout (Pwn Request pattern) Malicious npm lifecycle scripts (postinstall curl | sh, node -e inline exec) Hidden Unicode in AI config files (.cursorrules, .github/copilot-instructions.md) Wildcard tool permissions (Bash(), Edit()) Exposed .env files without .gitignore protection Cache poisoning + credential exposure combos Outputs text, JSON, or SARIF (for GitHub Code Scanning integration).

56 end-to-end tests. Every detection maps to a documented real-world attack.

MIT licensed. No signup, no paywall.

GitHub: https://github.com/aidevshield/aidevshield

Happy to answer questions about the threat landscape or implementation.