create-disposition

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill adheres to security best practices, communicating exclusively with the vendor's official API (api.bolna.ai) and requiring no external packages or untrusted dependencies.
  • [CREDENTIALS_UNSAFE]: Authentication is managed through the BOLNA_API_KEY environment variable. No hardcoded secrets, tokens, or sensitive keys were identified in the instructions, scripts, or asset files.
  • [PROMPT_INJECTION]: The skill involves processing transcripts using an LLM to extract structured data (Ingestion points: assets/sample_transcripts.json and inputs to scripts/test_dispositions.py; Boundary markers: absent; Capability inventory: limited to vendor API requests; Sanitization: absent). This surface is deemed safe because the extraction results do not drive high-risk secondary actions like shell execution or file system modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 01:54 PM
Security Audit — agent-trust-hub — create-disposition