rust-crate-discovery
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill performs network requests to well-known technology services including crates.io, GitHub, docs.rs, and RustSec. These requests are used to retrieve public metadata, documentation status, and security advisories, which is essential for the skill's primary function of dependency evaluation.
- [COMMAND_EXECUTION]: The skill executes a bundled Python utility located at
scripts/crate_eval.py. This script is implemented using only the Python standard library (e.g., urllib, json) and does not perform any unsafe operations such as subprocess spawning or arbitrary command execution. - [DATA_EXFILTRATION]: The tool transmits crate names to external APIs to fetch relevant metadata. This behavior is disclosed in the documentation and is restricted to public registry identifiers, posing no risk of sensitive data exposure.
- [PROMPT_INJECTION]: The skill processes and displays user-controlled metadata (such as crate descriptions and README links) from the crates.io registry. This constitutes a surface for indirect prompt injection; however, the tool's limited capabilities (read-only analysis) and use of structured JSON parsing minimize the practical risk.
Audit Metadata