agentic-engineering-github-tape-review
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI tool (gh) to execute queries against the GitHub API for pull requests, issues, and commit data.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and processes untrusted data from GitHub. 1. Ingestion points: Pull request titles, descriptions, and comments retrieved via gh commands in SKILL.md. 2. Boundary markers: Absent; there are no instructions to delimit or ignore instructions found within the retrieved GitHub data. 3. Capability inventory: Limited to fetching data via gh and providing structured analysis. 4. Sanitization: Absent; the skill does not perform validation or filtering of the content retrieved from GitHub.
Audit Metadata