binlee-med-aesthetics-strategy
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a step to execute a local Node.js script (
node ../binlee-source-library/scripts/search-corpus.mjs) to query a medical corpus. This is a functional requirement for its stated purpose and relies on local resources rather than untrusted remote sources. - [PROMPT_INJECTION]: An indirect prompt injection surface is present because the skill ingests data from local files and search results without defining explicit boundary markers or sanitization logic. This represents a standard risk for skills that process text data, but no malicious intent or active injections were detected.
- Ingestion points: The file
../binlee-source-library/references/evidence-policy.mdand the output from thesearch-corpus.mjsscript. - Boundary markers: No explicit delimiters (e.g., XML tags or backticks with warnings) are specified in the workflow to separate data from instructions.
- Capability inventory: File system read access and the ability to execute a local Node.js script.
- Sanitization: The instructions do not define methods for filtering or escaping content retrieved from the corpus.
Audit Metadata