stress-testing
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE] (SAFE): This skill is entirely composed of Markdown text. No executable scripts, binaries, or package manifests were found in the provided file.
- [Indirect Prompt Injection] (LOW): This skill identifies a vulnerability surface for indirect prompt injection. 1. Ingestion points: The skill is designed to process untrusted user-provided plans and web content using Read and WebFetch tools. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands within ingested data are defined. 3. Capability inventory: The agent is granted access to file system tools (Read, Glob, Grep) and web search tools (WebFetch, WebSearch). 4. Sanitization: There is no evidence of input validation, escaping, or sanitization of external content before it is processed by the LLM.
Audit Metadata