poke-integrations

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill documents the integration of external data sources such as GitHub, Gmail, and Sentry. This creates a surface for indirect prompt injection where malicious content in external data (like a PR description or email) could attempt to influence the agent's behavior.
  • Ingestion points: Data ingested from services like GitHub, Gmail, Outlook, Notion, and Sentry as described in references/builtin-integrations.md.
  • Boundary markers: None explicitly defined in the provided instructional text.
  • Capability inventory: The agent can perform actions via the Poke SDK and integrated tools.
  • Sanitization: No specific sanitization methods for external data are described in these reference files.
  • [COMMAND_EXECUTION]: The documentation describes the use of the vendor's CLI tool (poke) to manage integrations. This is a standard and expected administrative feature of the platform.
  • [CREDENTIALS_UNSAFE]: The skill mentions the use of API keys and OAuth for authentication. Examples use safe placeholders (e.g., sk-xxx), and the documentation specifies that keys are stored encrypted by the service.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 07:42 AM