geofeed-tuner

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the agent to generate and execute Python scripts for processing CSV data, performing network requests to an MCP server, and rendering HTML reports.
  • [EXTERNAL_DOWNLOADS]: It provides functionality to download geolocation feed files from remote URLs provided by the user for tuning and validation.
  • [DATA_EXFILTRATION]: IP prefixes and geographical strings are transmitted to the vendor's API (mcp.fastah.ai) to retrieve location refinements. This behavior is documented and aligns with the skill's primary function.
  • [PROMPT_INJECTION]: The skill processes untrusted input from external CSV files, posing an indirect prompt injection risk. Ingestion points: Phase 2 fetches data from local or remote CSV sources. Boundary markers: Not present for the input data. Capability inventory: The skill utilizes script execution, network access, and report generation. Sanitization: The instructions explicitly require the agent to HTML-escape all values during report generation to prevent injection vulnerabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 07:49 AM
Security Audit — agent-trust-hub — geofeed-tuner