github-issues
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the
ghCLI and MCP tools to perform repository management tasks. These operations are restricted to the GitHub API and are consistent with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: References external assets and raw content from
github.com. These targets are well-known, trusted services and the references are used for standard image embedding and documentation fetching. - [INDIRECT_PROMPT_INJECTION]: The skill includes tools to read issue comments and bodies (
mcp__github__issue_read). While this introduces a potential attack surface for indirect prompt injection from external contributors, the skill provides structured templates inreferences/templates.mdto maintain focus on specific tasks, reducing the risk of accidental obedience to embedded instructions. - [DYNAMIC_EXECUTION]: The
references/images.mdfile provides a code template for usingpuppeteer-coreto generate screenshots. This is a local automation task for the agent to assist the user and does not involve executing untrusted remote code.
Audit Metadata