push
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local development commands including
make -C elixir allandmix pr_body.check. These tools run project-specific scripts and code as part of the validation process. - [INDIRECT_PROMPT_INJECTION]: The skill fetches the body of a pull request from GitHub using
gh pr view. This external content is then processed by the agent to update or validate the pull request, creating an attack surface where instructions embedded in the PR body could influence agent behavior. - Ingestion points: The pull request body is read into a temporary file using
gh pr view --json body -q .bodyinSKILL.md. - Boundary markers: Absent; there are no specific delimiters or instructions to the agent to ignore embedded commands in the ingested text.
- Capability inventory: The skill has the capability to push code (
git push), create/edit PRs (gh pr create/edit), and execute build tasks (make,mix). - Sanitization: Absent; the external content is written to a file and processed without explicit sanitization or filtering.
Audit Metadata