lindy-core-workflow-a
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of AI agents by processing user-provided natural language instructions, which are then interpolated into agent specifications.
- Ingestion points: User definitions for agent 'Purpose' and 'Identity' in SKILL.md and the AgentSpec object in references/implementation-guide.md.
- Boundary markers: The prompt templates provided do not include specific delimiters or 'ignore' instructions to help the generated agents distinguish between instructions and potentially malicious data.
- Capability inventory: Lindy agents created using this workflow can execute code in sandboxed environments and perform network operations such as HTTP requests and email.
- Sanitization: There is no evidence of input validation or content sanitization for the instructions provided by the user before they are passed to the API.
- [EXTERNAL_DOWNLOADS]: The implementation guide specifies the use of the SDK for the Lindy AI platform to facilitate agent creation.
- Evidence: The skill utilizes the '@lindy-ai/sdk' package in references/implementation-guide.md.
Audit Metadata