twilio-enterprise-knowledge
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation recommends installing official vendor libraries using
pip install twilioandnpm install twilio. - [INDIRECT_PROMPT_INJECTION]: The skill implements a Retrieval-Augmented Generation (RAG) architecture that incorporates external data into the AI agent's prompt.
- Ingestion points: Content is retrieved from external web sources and PDFs via the Twilio Search API as documented in
SKILL.md. - Boundary markers: The provided prompt templates use headers like
Relevant knowledge:andDocumentation:to separate retrieved data from instructions. - Capability inventory: Knowledge chunks are concatenated into system prompt strings for the LLM.
- Sanitization: No explicit sanitization or filtering of the retrieved content is described before it is injected into the prompt.
- [COMMAND_EXECUTION]: The code examples in the skill perform network requests to official vendor domains (
knowledge.twilio.comandmemory.twilio.com) and read local files for content ingestion.
Audit Metadata