durability
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest and analyze project-specific instructions and source code, creating a surface for potential indirect prompt injection attacks from untrusted codebases.\n
- Ingestion points: Ingests project instructions, service definitions, and storage adapters as described in the 'Read the actual boundary' section of SKILL.md.\n
- Boundary markers: No explicit instructions are provided to the agent to distinguish between trusted guidance and untrusted content within the analyzed files.\n
- Capability inventory: The agent has the capability to modify source code and execute shell commands via 'pnpm turbo'.\n
- Sanitization: There are no documented steps for sanitizing or validating external input before processing.\n- [COMMAND_EXECUTION]: The skill requires the agent to run project-level scripts which could be leveraged if the environment is compromised.\n
- Evidence: Instructions to execute 'pnpm turbo run check test build' in SKILL.md.\n- [EXTERNAL_DOWNLOADS]: The skill references external technical resources and the author's public repository for configuration and references.\n
- Evidence: Multiple links to 'github.com/joelhooks/rat-stack' and technical publisher websites in references/bibliography.md.
Audit Metadata