Scan a GitHub repository for skill files and security issues.
AI agents call scan_github to retrieve information from SkillAudit without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
repo | string | — | GitHub repository in owner/name format (e.g. modelcontextprotocol/servers) |
branch | string | — | Branch to scan (default: main) |
Parameters from the server's own tool schema.
scan_github retrieves and analyzes repository contents to identify security issues. This is fundamentally a Read operation—it queries/inspects a GitHub repository. However, severity is elevated to 'high' because scanning repositories may expose sensitive data like credentials, tokens, or proprietary code that an AI agent could exfiltrate or misuse if the tool's output is not properly sandboxed.
From the tool's definition Tool description states 'Scan a GitHub repository for skill files and security issues.' The verb 'scan' and 'detect' indicate data retrieval and analysis without modification.
Documented attack patterns abuse exactly the kind of access scan_github gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and SkillAudit, and nothing reaches the server without passing your rules. This is the rule we recommend for scan_github:
{
"version": "1",
"default": "deny",
"tools": {
"scan_github": {}
}
}scan_github is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Scan a GitHub repository for skill files and security issues. It is categorised as a Read tool in the SkillAudit MCP Server, which means it retrieves data without modifying state.
scan_github accepts 2 parameters: repo, branch. The full parameter table on this page comes from the server's own tool schema.
Register the SkillAudit MCP server in PolicyLayer and add a rule for scan_github: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches SkillAudit. Nothing to install.
scan_github is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the scan_github rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for scan_github. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
scan_github is provided by the SkillAudit MCP server (megamind-0x/skillaudit). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from SkillAudit, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
4 SkillAudit tools catalogued and risk-classified — across an index of 43,000+ MCP servers.