find-editorial-citations

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for using the citeguild command-line interface and standard curl commands to interact with the service.
  • Evidence: Examples include citeguild search --json and curl --fail-with-body https://citeguild.lvtd.dev/api/v1/search.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with an external API hosted at citeguild.lvtd.dev to fetch search results.
  • Evidence: The REST API transport targets https://citeguild.lvtd.dev/api/v1/search.
  • [CREDENTIALS_SAFE]: The skill includes explicit instructions for the secure handling of the CITEGUILD_API_KEY, warning against hardcoding, logging, or passing credentials in cleartext parameters.
  • Evidence: The 'Authentication and errors' section mandates the use of environment variables and warns: "Never hardcode, print, log, or commit a credential."
  • [DATA_EXPOSURE]: The workflow emphasizes privacy by instructing the agent to extract focused search queries and avoid sending confidential or unnecessary text to the external API.
  • Evidence: Step 1 of the workflow states: "Preserve the user's meaning and avoid sending confidential or unnecessary text."
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:51 AM
Security Audit — agent-trust-hub — find-editorial-citations