fp-async
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing data from external sources like APIs and the local file system, which creates a surface area for indirect prompt injection if the ingested data is attacker-controlled.
- Ingestion points: Data is ingested via
fetchAPI responses andfs.readFileoperations (SKILL.md, references/detailed-guide.md). - Boundary markers: The instructions recommend validating unknown JSON against a runtime schema before use, providing a layer of protection against unexpected data structures.
- Capability inventory: The code examples demonstrate network requests (
fetch), database operations (PrismaClient), and file system interactions (fs.readFile,fs.writeFile). - Sanitization: The patterns include the use of
encodeURIComponentfor URL parameters and advise against serializing raw database or system errors to public responses.
Audit Metadata