bilingual-doc-sync
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The
scripts/worldset_coverage.pyutility invokes the GitHub CLI (gh) usingsubprocess.run. This is a functional requirement for the skill's expansion audit feature, which measures language coverage across public repositories. The command is executed using a list of arguments rather than a single string, preventing shell injection vulnerabilities. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted document content during synchronization and audits.
- Ingestion points: Reads content from local Markdown and HTML files via
scripts/check_language_parity.pyand fetches repository metadata and file trees viascripts/worldset_coverage.py. - Boundary markers: The workflow utilizes
<!-- lang-only -->HTML comments to scope and isolate language-specific content. - Capability inventory: Performs local file reads and executes the
ghCLI for remote repository inspection. - Sanitization: Uses specific regular expressions to extract links, which focuses the agent's attention on non-executable structural elements and largely ignores arbitrary prose.
- [SAFE]: No obfuscation, persistence mechanisms, or unauthorized data exfiltration attempts were found. The skill appropriately manages its dependencies and interactions with the GitHub API using standard, non-malicious patterns.
Audit Metadata