v0.1
Getting Started
Install and run your first scan with skill-issue
Getting Started
skill-issue is a static security analyzer for Claude skill directories. It detects security vulnerabilities, prompt injection risks, malicious code patterns, and configuration issues in AI skill definitions.
Installation
Download the binary
Grab the latest release for your platform from GitHub Releases:
# macOS (Apple Silicon)
curl -L https://github.com/daviddrummond95/skill-issue-cli/releases/latest/download/skill-issue-aarch64-apple-darwin -o skill-issue
chmod +x skill-issue
sudo mv skill-issue /usr/local/bin/
# macOS (Intel)
curl -L https://github.com/daviddrummond95/skill-issue-cli/releases/latest/download/skill-issue-x86_64-apple-darwin -o skill-issue
chmod +x skill-issue
sudo mv skill-issue /usr/local/bin/
# Linux (x86_64)
curl -L https://github.com/daviddrummond95/skill-issue-cli/releases/latest/download/skill-issue-x86_64-unknown-linux-gnu -o skill-issue
chmod +x skill-issue
sudo mv skill-issue /usr/local/bin/
Build from source
If you have Rust installed, you can build from source:
git clone https://github.com/daviddrummond95/skill-issue-cli.git
cd skill-issue-cli
cargo build --release
The compiled binary will be at target/release/skill-issue.
Your first scan
Run skill-issue against any skill directory:
# Scan the current directory
skill-issue
# Scan a specific skill directory
skill-issue /path/to/my-skill
You'll see output like this:
╭──────────┬────────────┬──────────────┬──────┬──────────────────────────────────────╮
│ Severity │ Rule │ File │ Line │ Message │
├──────────┼────────────┼──────────────┼──────┼──────────────────────────────────────┤
│ ERROR │ SL-SEC-001 │ README.md │ 24 │ Possible API key detected │
│ WARNING │ SL-HID-002 │ README.md │ 12 │ HTML comment may hide instructions │
╰──────────┴────────────┴──────────────┴──────┴──────────────────────────────────────╯
Found 2 issue(s): 1 error(s), 1 warning(s), 0 info(s)
CLI Options
| Option | Short | Description | Default |
|---|---|---|---|
--format |
-f |
Output format: table, json, or sarif |
table |
--config |
-c |
Path to configuration file | Auto-detect |
--severity |
-s |
Minimum severity: info, warning, error |
info |
--ignore |
Rule IDs to ignore (repeatable) | None | |
--error-on |
Minimum severity for non-zero exit | error |
|
--quiet |
-q |
Suppress output except findings | Off |
--verbose |
-v |
Show detailed rule information | Off |
--no-color |
Disable colored output | Off |
Exit Codes
| Code | Meaning |
|---|---|
0 |
Clean scan (or only findings below --error-on threshold) |
1 |
Warnings found (when --error-on warning) |
2 |
Errors found or threshold exceeded |
Next Steps
- Configuration — Customize rules and thresholds
- Rules Reference — Browse all 50+ security rules
- CI/CD Integration — Automate scanning in your pipeline