core-dynamic-skills
Pass
Audited by Gen Agent Trust Hub on Jun 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Retrieves crate documentation and module details from docs.rs, a well-known service for Rust project documentation.
- [COMMAND_EXECUTION]: Uses standard Unix utilities such as cat and grep to extract dependency information from Cargo.toml files, and performs file system operations (mkdir, rm) within ~/.claude/skills/.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes untrusted documentation from docs.rs to generate agent instructions.
- Ingestion points: Cargo.toml and external documentation from docs.rs.
- Boundary markers: None specified for protecting generated SKILL.md content from instructions embedded in the documentation.
- Capability inventory: File read/write via shell commands, web browsing via agent-browser and WebFetch.
- Sanitization: No sanitization or validation of the fetched documentation content is described.
Audit Metadata