doc-lookup
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js helper script to perform searches and fetch data.
- Evidence:
node "$SKILL_DIR/scripts/doc-lookup.cjs" searchandgetcommands described inSKILL.md. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves content from external documentation APIs (Context Hub and Context7) and incorporates it into the agent's context, creating a potential attack surface if those sources were compromised.
- Ingestion points: Output from the
doc-lookup.cjsscript is used as evidence for the agent's reasoning inSKILL.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the fetched documentation as untrusted data.
- Capability inventory: The agent is instructed to use these findings to verify security claims (e.g., SQL injection protections, input validation) which directly influences the agent's logic for bug fixing.
- Sanitization: No validation or sanitization of the external documentation content is mentioned before it is processed by the LLM.
Audit Metadata