web-research-parallel

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill directs the agent to read credentials from the local file path ~/.gooseworks/credentials.json.
  • [COMMAND_EXECUTION]: The skill uses inline Python commands to parse local configuration files and export environment variables during the setup process.
  • [DATA_EXFILTRATION]: Local API credentials are transmitted to the gooseworks.ai API endpoint, which is a resource owned by the vendor.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection. Ingestion points: Data is retrieved from the web via search and extraction endpoints in SKILL.md. Boundary markers: Absent. Capability inventory: The skill can perform network operations and API calls using curl. Sanitization: There is no evidence of content filtering or validation for external web data.
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to install and execute the goose-skills and gooseworks packages from the npm registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 06:18 AM
Security Audit — agent-trust-hub — web-research-parallel