vinta-analyze-codebase
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize large amounts of untrusted content from a repository, which may contain instructions targeting the AI orchestrator or downstream skills.
- Ingestion points: Numerous repository files including
package.json,pyproject.toml,README.md,AGENTS.md,CLAUDE.md, test files, and CI configuration files. - Boundary markers: The skill does not define specific delimiters or "ignore instructions" wrappers for the content it aggregates into the final structured inventory.
- Capability inventory: The skill uses
ls,cat,grep, andgitfor repository analysis. It does not have file-write or network-access capabilities. - Sanitization: There is no evidence of content filtering or escaping for the raw text captured from repository documentation (e.g.,
AGENTS.md) before it is passed to the orchestrator. - [COMMAND_EXECUTION]: The skill uses local shell commands to inspect the repository environment.
- Evidence: Usage of
ls -la,git log,git symbolic-ref, andgit remote get-url. These are read-only operations restricted to repository metadata and file listing.
Audit Metadata