create-rfc
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and adapts an RFC structure from the official Rust language repository on GitHub.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as repository code, documentation, and user-supplied notes to research and develop the RFC, which presents a surface for indirect prompt injection.
- Ingestion points: The agent reads supplied materials, relevant repository guidance, existing RFC conventions, and primary code sources as outlined in
SKILL.md. - Boundary markers: The instructions direct the agent to distinguish between "source-supported capability, observed behavior, proposed design, and unverified integration," which acts as a logical boundary for processing information.
- Capability inventory: The skill is authorized to read files, write to the
docs/rfcs/directory, and perform small, reversible local experiments subject to user authorization. - Sanitization: There are no specific instructions for sanitizing or escaping the content read from external sources.
Audit Metadata