Skills
Navigate
skills/deepgram/deepgram-python-sdk/deepgram-python-text-intelligence/Gen Agent Trust Hub

deepgram-python-text-intelligence

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing deepgram/skills via npx. This is an official vendor command for extending agent functionality.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates processing external text and URLs, which is an inherent part of its text analysis functionality. This represents a potential surface for indirect prompt injection from untrusted data. 1. Ingestion points: SKILL.md (via text and url fields in the request dictionary). 2. Boundary markers: Not included in the instructional snippets. 3. Capability inventory: SDK methods that perform authenticated network requests to Deepgram's API. 4. Sanitization: Not explicitly demonstrated in the quickstart examples.- [SAFE]: The skill uses best practices for secret management by demonstrating the usage of python-dotenv to avoid hardcoding API keys.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 11:24 AM
Security Audit — agent-trust-hub — deepgram-python-text-intelligence