entity-optimizer

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python scripts found in the ${CLAUDE_PLUGIN_ROOT} to fetch entity data.
  • Evidence: Commands such as python3 "${CLAUDE_PLUGIN_ROOT}/scripts/connectors/kg.py" reconcile "<entity>" are used to interface with the Knowledge Graph API.
  • These operations are restricted to internal helper scripts and are used solely for the skill's primary purpose of entity optimization.
  • [PROMPT_INJECTION]: The skill processes information from external sources (Wikidata, Wikipedia, GDELT), creating an attack surface for indirect prompt injection.
  • Ingestion points: Results from the Knowledge Graph API and GDELT news mentions are incorporated into the agent's context during audits.
  • Boundary markers: Explicit boundary markers for external data are not specified in the instructions.
  • Capability inventory: The skill can execute local scripts and write entity profiles to the memory/entities/ folder.
  • Sanitization: Data from well-known and reputable services (Wikidata, GDELT) is ingested to provide entity health scores and signal analysis. This is a low-risk pattern given the trusted nature of the data sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 10:02 AM
Security Audit — agent-trust-hub — entity-optimizer