felo-livedoc
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute a local JavaScript file (
scripts/run_livedoc.mjs) which handles all communication with the Felo API. - [EXTERNAL_DOWNLOADS]: The skill connects to the vendor's official API at
https://openapi.felo.ai. This behavior is expected as it facilitates the skill's primary functions using the author's own services. - [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involving the ingestion of external URL content and file data creates an indirect prompt injection surface.
- Ingestion points: Data enters the agent's context through the
add-urls,upload,retrieve, andcontentcommands. - Boundary markers: The skill does not provide specific instructions or delimiters to the agent to distinguish between its own instructions and untrusted content retrieved from the knowledge base.
- Capability inventory: The skill has the capability to read local files for upload and perform network requests to the vendor API using the
fetchAPI. - Sanitization: There is no evidence of sanitization or validation of the content retrieved from external sources before it is processed by the agent.
Audit Metadata