ai-csv-excel-merger

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to process user data through the AI Skills API. All network communications in scripts/run.py are directed to the vendor's official domain (ai-skills.ai).
  • [PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources (text input, file uploads, and external URLs) defined in references/form-schema.json. This constitutes an indirect prompt injection surface where malicious instructions within the processed documents could attempt to influence the agent's output.
  • Ingestion points: Untrusted data enters the context via materialText, materialFile, and materialUrl in references/form-schema.json.
  • Boundary markers: Not explicitly defined in the skill instructions; processing is handled by the remote API.
  • Capability inventory: The associated runner script scripts/run.py is limited to network requests to the vendor API and does not perform file system writes or subprocess execution based on input.
  • Sanitization: Content is transmitted to the AI Skills platform which is responsible for sanitization and safety filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:30 AM
Security Audit — agent-trust-hub — ai-csv-excel-merger