azure-cosmos-ts

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill enables the ingestion of external data from Azure Cosmos DB, which introduces a potential surface for indirect prompt injection if the database contains malicious instructions intended to manipulate the agent.
  • Ingestion points: Data is ingested via container.items.query() and container.item().read() calls as documented in SKILL.md.
  • Boundary markers: The skill does not provide explicit instructions for the agent to use delimiters or specific safety prompts when processing content returned from the database.
  • Capability inventory: The skill utilizes the @azure/cosmos SDK to perform network-based CRUD operations.
  • Sanitization: The documentation correctly recommends using SqlQuerySpec for parameterized queries to prevent SQL injection within the database itself.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install official packages from a well-known service provider.
  • Evidence: The skill specifies npm install @azure/cosmos @azure/identity in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 01:54 AM
Security Audit — agent-trust-hub — azure-cosmos-ts