status-report
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external, potentially untrusted sources such as pull request titles, descriptions, and review comments, which could contain instructions meant to influence the agent's summary or behavior.
- Ingestion points: Steps 2 and 3 of the 'Gather evidence' section in
SKILL.mdinvolve reading data from PRs, review threads, and CI summaries. - Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the gathered evidence.
- Capability inventory: The skill uses
gitandgh(GitHub CLI) via shell execution to gather metadata and repository state. - Sanitization: Absent; the instructions do not specify any validation or escaping of the ingested text before it is summarized or rendered into the board.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands using
gitandghto enumerate worktrees, branches, and pull request states. While these are standard tools for the stated purpose, they represent the skill's primary interface with the host system.
Audit Metadata