effect
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill explicitly instructs the agent to read and follow instructions contained in
node_modules/effect/AGENTS.md. This behavior creates an attack surface where potentially malicious or adversarial instructions embedded within the package's documentation could be executed by the agent without further verification. - Ingestion points: Content is ingested from
node_modules/effect/AGENTS.md,node_modules/effect/ai-docs/, and source files innode_modules/effect/srcandnode_modules/@effect/*. - Boundary markers: The skill lacks explicit boundary markers or instructions to treat the external file content as untrusted data.
- Capability inventory: The agent utilizing this skill typically possesses file system read/write access and the ability to execute code or shell commands.
- Sanitization: There is no evidence of sanitization or filtering of the instructions found within the referenced documentation files before the agent is told to follow them.
Audit Metadata