engineering-retro
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes git commit history and pull request data which are external, potentially untrusted sources of information.
- Ingestion points: The skill reads git commit messages and author names (Step 2) as well as GitHub Pull Request titles and review metadata (Step 8).
- Boundary markers: Absent. The agent processes the raw output of shell commands (git log, gh pr list) without using specific delimiters or instructions to ignore embedded instructions within the commit data.
- Capability inventory: The skill has the ability to execute shell commands (git, gh, sed, awk) and perform local file writes (creating JSON snapshots and modifying .gitignore).
- Sanitization: Absent. There is no evidence of filtering or escaping logic for the ingested strings before they are incorporated into the narrative summary (Step 13).
- [COMMAND_EXECUTION]: The skill relies on the execution of multiple shell-based tools to perform its analysis.
- Evidence: Execution steps involve git log for commit history, gh pr list for pull request metrics, and utilities like sed, awk, and grep for data parsing. These are standard developer tools used for their intended purpose.
Audit Metadata