Skills
skills/full-stack-skills/rust-skills/rust-http-client/Gen Agent Trust Hub

rust-http-client

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to fetch and process data from arbitrary external URLs.
  • Ingestion points: The HttpClient::get_text method in examples/golden-http-client/src/lib.rs and the execution scenarios in examples/examples.md describe fetching remote content based on user-provided or external locations.
  • Boundary markers: The instructions in SKILL.md (Section 5, 'Enforce trust boundaries') explicitly advise developers to restrict schemes, hosts, and ports for user-controlled destinations to prevent SSRF.
  • Capability inventory: The skill provides instructions for utilizing the reqwest library for network operations and includes logic for local integration testing using TcpListener.
  • Sanitization: The reference code in lib.rs demonstrates safe data handling by enforcing strict limits on the number of bytes read from a response body using the .take() method, protecting against denial-of-service through resource exhaustion.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 06:56 PM
Security Audit — agent-trust-hub — rust-http-client