speed-cameras

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill's behavior aligns with its documented purpose of helping users locate speed cameras.- [COMMAND_EXECUTION]: Utilizes a dynamic context injection check in SKILL.md to confirm the installation status of the uv tool. This is a benign environment check and does not involve user-supplied input.- [EXTERNAL_DOWNLOADS]: Fetches geolocation data and address information from nominatim.openstreetmap.org and ip-api.com. These are well-known services necessary for the skill's geocoding functionality.- [PROMPT_INJECTION]: Evaluated for indirect prompt injection surfaces as the skill ingests data from external APIs.
  • Ingestion points: Address and location details fetched via httpx in scripts/speed_cameras.py.
  • Boundary markers: Results from external APIs are parsed as structured JSON; no explicit instruction-delimiting markers are used.
  • Capability inventory: The script performs network requests and maintains a local cache in ~/.config/speed-cameras/.
  • Sanitization: Data from external sources is mapped to specific result fields in a JSON object and is not used to dynamically construct agent instructions or shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 01:33 AM
Security Audit — agent-trust-hub — speed-cameras