adopt-project
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the user's local repository, which could contain malicious instructions designed to influence the agent's behavior during the triage or planning phases.
- Ingestion points: The skill reads project-level metadata and documentation, including
package.json,READMEfiles, and previous artifacts in thelearning/directory. - Boundary markers: There are no explicit instructions to use boundary markers or delimiters when interpolating content from the codebase into the agent's context.
- Capability inventory: The skill allows the agent to read and write files within the workspace and execute command-line utilities (
git,gh). - Sanitization: The skill mitigates risks by requiring the agent to verify all "load-bearing claims" against the actual state of the repository rather than relying on potentially manipulated documents or agent memory.
- [COMMAND_EXECUTION]: The skill utilizes local command-line tools to perform project inventory and verification.
- Evidence: The instructions explicitly direct the agent to use
git status,git log, and the GitHub CLI (gh) to determine the project's version control state and verify if branches or PRs are merged.
Audit Metadata