space-earth-science-explorer

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to download and install setup dependencies (such as 'what-is-agentpmt' and 'agentpmt-account-mcp-rest-api-setup') from the AgentPMT/agent-skills repository via the 'npx skills' and 'openclaw' commands.
  • [COMMAND_EXECUTION]: Provides standard setup commands for terminal execution to initialize the skill's environment and connect to the AgentPMT API service.
  • [DATA_EXFILTRATION]: The tool transmits user queries to the 'api.agentpmt.com' endpoint for processing and retrieves scientific data. This behavior is consistent with the skill's stated purpose as a remote tool connector.
  • [PROMPT_INJECTION]: Indirect surface exists as the skill ingests external data (descriptions and titles from NASA/NOAA/USGS). However, the skill treats this as structured data for research and does not provide an execution path for instructions embedded in the results.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 08:14 AM