wiki-manager
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard GitHub and Git tools including
gh apiandgit(clone, commit, push) to interact with wiki repositories. These operations are transparently defined and essential for the skill's primary function of wiki management. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to ingest and process content from external Git repositories (GitHub Wikis).
- Ingestion points: Content is retrieved via
git clonefrom the target repository's wiki git URL ({repo}.wiki.git). - Boundary markers: No explicit boundary markers or instructional delimiters are defined in the SKILL.md to protect against instructions embedded within wiki markdown files.
- Capability inventory: The skill possesses capabilities to execute shell commands (
gh,git), read/write local files, and push changes to remote repositories. - Sanitization: No specific sanitization or validation logic is specified for the markdown content prior to the agent processing it.
- [SAFE]: The overall logic of the skill is consistent with its stated purpose. It does not contain any detected obfuscation, hardcoded credentials, unauthorized network requests, or attempts at persistence or privilege escalation.
Audit Metadata