ci-triage
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs legitimate CI triage activities using GitHub integration tools. It follows a structured taxonomy to classify failures and maintains a single issue for main branch failures, which is standard development practice. No suspicious command execution, persistence, or data exfiltration behaviors were found.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes untrusted data from the repository. Evidence:
- Ingestion points: The skill reads commit messages, pull request details, and workflow job metadata using GitHub MCP tools (SKILL.md).
- Boundary markers: The skill does not define specific delimiters to isolate external repository data from its own instructions.
- Capability inventory: The skill can create and update GitHub issues and comments using GitHub MCP tools (SKILL.md).
- Sanitization: No explicit logic is provided to sanitize or escape data retrieved from the repository before use.
Audit Metadata