issue-triage
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches configuration for the GitHub MCP server from the official GitHub Copilot API endpoint.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-generated content from GitHub issues, creating an attack surface for indirect prompt injection where instructions embedded in issue bodies could attempt to manipulate the triage process.
- Ingestion points: Data enters the context via GitHub issue fetching commands (
list_issuesorgh issue list) described in the 'Workflow' section of SKILL.md. - Boundary markers: The skill does not define specific boundaries or "ignore" instructions for the content being triaged.
- Capability inventory: The skill instructions encompass write access to the repository, including labeling, commenting, and closing issues as described in the 'Integration' section of SKILL.md.
- Sanitization: No filtering or validation of issue body content is specified before the agent analyzes it.
Audit Metadata