vs-grill-me
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted repository content. Ingestion points include plans, PRDs, and codebase files read during the 'Pre-scan' phase in SKILL.md. There are no explicit boundary markers or sanitization procedures defined for this ingested content. The agent's capabilities while handling this data include the ability to write reports to the local file system and perform network searches using MCP tools.
- [EXTERNAL_DOWNLOADS]: The README.md file directs users to install the skill from the author's GitHub repository (github.com/vltansky/skills).
- [SAFE]: The skill requires extensive read access to the local codebase and documentation to provide context-aware feedback. These actions are legitimate and necessary for the skill's primary function of stress-testing development plans.
Audit Metadata