Skill Scan Gate — Development skill for Claude Code
CI gate for Claude Code skills and plugins: scans a skills or plugin repository for instruction-override text, exfiltration patterns, dangerous hook commands, unpinned MCP servers and secret-shaped st.
How to install Skill Scan Gate
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open basitalisandhu/skill-scan-gate and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Skill Scan Gate does
CI gate for Claude Code skills and plugins: scans a skills or plugin repository for instruction-override text, exfiltration patterns, dangerous hook commands, unpinned MCP servers and secret-shaped strings, writes SARIF for code scanning, and fails the build above your threshold. GitHub Action and CLI.
Alternatives in Development
- JWT Scan — JWT attack toolkit (offline) — alg:none forgery, RS256→HS256 algorithm confusion, weak-secret crack, static cl 4.5k ★
- Claude-code-safety-net Rules Reference — Custom rules reference for defining safe/dangerous command patterns 1.2k ★
- Pi Reserved Keybindings — Extensions cannot override these shortcuts — they are silently skipped by registerShortcut() 526 ★
README
skill-scan-gate: CI gate for Claude Code skills and plugins
**skill-scan-gate scans a skills or plugin repository in CI for instruction-override text, exfiltration patterns, dangerous hook commands, unpinned MCP servers and secret-shaped strings, writes SARIF for code scanning, and fails the build on findings above your threshold.**
[](https://github.com/basitalisandhu/skill-scan-gate/actions/workflows/ci.yml) [](https://github.com/basitalisandhu/skill-scan-gate/actions/workflows/self-test.yml) [](LICENSE) [](pyproject.toml)
GitHub Action:
- uses: basitalisandhu/skill-scan-gate@v0.1.0
pip (after the first PyPI release; until then `pipx install git+https://github.com/basitalisandhu/skill-scan-gate`):
pip install skill-scan-gate
Container image (GitHub Packages):
docker run --rm -v "$PWD:/work:ro" ghcr.io/basitalisandhu/skill-scan-gate:0.1.0 scan .
What it is, who it is for, and why
A Claude Code skill is Markdown the model reads as instructions. A plugin adds hooks that run shell commands on tool calls, MCP servers that start with the session, and more skills, commands and agents. All of it arrives through pull requests to a repository: a team's shared skills, a plugin marketplace, a project's `CLAUDE.md` and `.mcp.json`. A reviewer reading a 300-line skill diff can miss one sentence that tells the model to ignore earlier instructions, or a hook that pipes a download into a shell.
skill-scan-gate is for maintainers of skills repositories, plugin authors and marketplace owners, and for any team that keeps agent configuration in git and reviews it in pull requests. It runs the same chec
Related Skills
Narthex
Prompt-injection defenses for Claude Code. A PreToolUse Bash hook blocks compositional credential-exfiltration
Heeler Secrets Scan
Run Heeler secret scanning for a repository or staged changes. Use when comitting code or the user asks to det
PDF Text Router
A Claude Code hook that routes each Read on a PDF to the path that actually works: text layer to UTF-8 text, s
Prompt Injection Guard
Detect prompt injection, hardcoded secrets, destructive commands and exfiltration patterns in any text or file
Common
Hard constraints that override any other instruction in this prompt:
Secret Scan
Scan working directory untuk secrets — gitleaks detect mode, custom rules, fail-fast kalau ada leak.
Related Agents
Kenaz
Audits the security of Claude Code plugins, MCP servers, and AI agents before installation. Analyzes executabl
Marketing Privacy Reviewer
Read-only privacy scout for cmo-plugin footage reels. Deployed by /cmo:review before the owner watches. Scans
Redteam
Use for anything under redteam/ — building prompt-injection and exfiltration suites, mapping what an injected