prepare-branch-context
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
gitandgh(GitHub CLI) commands to fetch repository metadata, logs, diffs, and pull request information. These commands are necessary for the skill's primary purpose of analyzing branch context. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data, creating an attack surface where malicious instructions in a PR or code diff could influence follow-up agents.
- Ingestion points: Pull request titles and bodies (via
gh pr view) and source code content (viagit diff). - Boundary markers: The instructions do not mandate the use of delimiters or 'ignore' instructions when processing or passing this untrusted data to the
BRANCH_CONTEXT_SUMMARYblock. - Capability inventory: The skill uses shell commands for analysis and coordinates with subagents like
exploreto process results. - Sanitization: No specific sanitization, filtering, or validation is performed on the ingested PR or code content.
Audit Metadata