recall
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to interact with version control systems and external services, specifically
git,sl(Sapling), and thegh(GitHub CLI). These are used to retrieve branch status, commit history, and pull request details. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external and shared sources that could potentially contain malicious instructions.
- Ingestion points: The skill reads GitHub PR descriptions and reviews via
gh pr view, as well as commit messages and local vault files (vault/index.md). - Boundary markers: The instructions explicitly constrain the agent to a "Read-only" mode, stating the deliverable is a "brief, not a change," and forbidding unasked updates to status tags.
- Capability inventory: The skill is limited to reading files, executing information-retrieval CLI tools (
git,sl,gh), and providing prose summaries. It does not perform automated writes or network operations to unknown domains. - Sanitization: No explicit sanitization of the content fetched from GitHub or commit logs is mentioned.
Audit Metadata