navigating-github
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the user's local repository, which could be exploited to influence the agent's behavior.
- Ingestion points: The skill reads local files and Git history (e.g.,
git log,git status) as part of the curriculum defined inSKILL.mdandreferences/learning-curriculum.md. - Boundary markers: The skill does not specify the use of delimiters when interpolating data from the repository into the agent's prompt context.
- Capability inventory: The skill utilizes
WriteandBashtools (restricted to git, gh, ssh, test, and echo commands) to manage repositories and files. - Sanitization: There are no explicit instructions for sanitizing data ingested from the repository before it is processed by the agent, although the skill includes robust rules for avoiding the commitment of secrets to the repository history in
references/safety-rules.md.
Audit Metadata