what-would-lenny-do
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved via the
lennysdataMCP server tools. - Ingestion points: Newsletter and podcast content fetched via
search_contentandread_contentinSKILL.md. - Boundary markers: The instructions direct the agent to synthesize and apply the frameworks, but do not provide explicit delimiters for the raw tool output.
- Capability inventory: Limited to reading and searching content via the specified MCP tools.
- Sanitization: No specific filtering or sanitization of the retrieved archive content is performed.
- [SAFE]: The skill uses a scoped set of MCP tools (
lennysdata:search_content,lennysdata:read_content,lennysdata:read_excerpt,lennysdata:list_content) to perform its primary function. No unauthorized network activity, file system access, or obfuscated code was detected.
Audit Metadata