team-debug
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
git(diff, show) and the GitHub CLI (gh pr,gh issue) to gather context from the local repository and remote GitHub issues/pull requests. These commands are necessary for its stated purpose of bug analysis.- [EXTERNAL_DOWNLOADS]: The skill fetches metadata and content from GitHub via theghtool. GitHub is a well-known and trusted service for software development.- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8) as it processes data from external sources (issue bodies, PR comments, and code files). - Ingestion points: Data is ingested through
gh pr view,gh issue view, and filesystem reads usingGlob+Readpatterns. - Boundary markers: No specific delimiters or "ignore instructions" warnings are defined to separate untrusted data from the agent's instructions.
- Capability inventory: The lead agent spawns several
general-purposesubagents. These subagents typically have broad tool access within the agent's environment. - Sanitization: There is no evidence of sanitization, filtering, or validation of the ingested external content before it is processed by the model or subagents.
Audit Metadata