repo-sync-steward
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local command-line tools including
git(fetch, cherry-pick, worktree, push) andrg(ripgrep) to manage repository synchronization. It also executes local Node.js scripts (scripts/validate-blog-frontmatter.jsandscripts/generate_feed.mjs) to validate content integrity. - [EXTERNAL_DOWNLOADS]: Interacts with remote GitHub repositories (remotes 'origin' and 'vercel-website'). These resources are associated with the 'frankxai' organization and the skill's primary synchronization purpose.
- [PROMPT_INJECTION]: The workflow involves ingesting and processing untrusted data from the
content/blog/**path. - Ingestion points: Markdown files and content within the
content/blog/directory (SKILL.md). - Boundary markers: None specified to delimit content from instructions.
- Capability inventory: Subprocess execution for Git, Node.js, and ripgrep.
- Sanitization: Relies on a Node.js validation script, but lacks explicit sanitization for instructions embedded within blog content.
Audit Metadata