skills.sh
Sigil Skill
Eight-phase security scanning for every AI coding agent. One command to install. Works offline. Scans repos, packages, MCP servers, and installed skills before your agent executes them.
Install
Or for specific agents: npx skills add nomarj/sigil-skill -a claude-code -a cursor -a codex
Works with 12 agents
Any agent that supports skills.sh gets Sigil security scanning automatically.
Eight scan phases
Every scan runs the full Sigil analysis pipeline with weighted severity scoring.
Install Hooks
setup.py cmdclass, npm postinstall scripts, Makefile install targets
Code Patterns
eval, exec, pickle, child_process, dynamic code execution
Network / Exfil
Outbound HTTP calls, webhooks, DNS tunnelling, socket connections
Credentials
ENV var access, hardcoded keys, SSH/AWS credential files
Obfuscation
Base64 encoding, character code arrays, hex-encoded payloads
Provenance
Hidden files, binaries, shallow clones, suspicious filenames
Prompt Injection
Jailbreaks, instruction overrides, system prompt exfiltration, tool abuse
Skill Security
Malicious skill manifests, MCP server exploits, credential embedding
Capabilities
Scan on demand
Ask your agent to scan any target — directory, repo URL, package name, or GitHub shorthand. Just say "scan this repo" or "is owner/repo safe?"
Pre-clone quarantine
Before cloning a repo or installing a package, the skill intercepts, scans in quarantine, and presents findings before any code reaches your environment.
Environment audit
Scans your local environment for exposed credentials — .env files with API keys, SSH keys with wrong permissions, secrets leaked in shell history.
Installed skills audit
Scans all skills installed across all your agent directories and reports risk assessments for each.
How it works
Install the skill
One npx command. The Sigil CLI binary is downloaded automatically on first use.
Ask your agent to scan
"Scan this repo", "is this package safe?", "audit my environment" — natural language.
Review the verdict
The skill returns a risk verdict (LOW / MEDIUM / HIGH / CRITICAL) with detailed findings grouped by phase.
Decide
Your agent presents findings and waits for your approval before proceeding with any installation or execution.
Sigil provides risk assessments based on automated pattern detection. Risk assessments do not constitute a guarantee of security or a definitive determination of malicious intent. This is not a replacement for server-side scanning — it's an additional layer that catches what server-side providers cannot.
SIGIL by NOMARK