fcode-ama
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documents the installation and use of the
@factorialco/fcode-clitool via package managers likepnpm. It provides standard commands for cloning workspaces and running processes, which are intended for local developer use within the platform's ecosystem. - [INDIRECT_PROMPT_INJECTION]: The instructions direct the agent to fetch and read live documentation from
https://code.factorialhr.comto provide accurate support. While this creates an ingestion point for external data, the source is the official vendor documentation site. - Ingestion points: Public documentation URLs under the
code.factorialhr.comdomain. - Boundary markers: There are no explicit instructions to the agent to treat fetched documentation as untrusted or to use specific delimiters.
- Capability inventory: The skill itself is informational and lacks the capability to write to the file system or perform unauthorized network transfers.
- Sanitization: No sanitization of the documentation content is defined, as the source is considered authoritative and trusted.
Audit Metadata