connect

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install the composio Python package and the @composio/core Node.js package from official package registries to enable the tool routing functionality.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of data from numerous external applications, which introduces an attack surface for indirect prompt injection via untrusted content from those services.
  • Ingestion points: Data retrieved from third-party integrations (e.g., GitHub issues, Gmail messages, Slack posts) via the Composio Tool Router.
  • Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore' instructions for content retrieved from external tools.
  • Capability inventory: The agent is granted extensive capabilities to write data and perform actions (e.g., sending emails, posting messages, creating issues) based on processed inputs.
  • Sanitization: The provided instructions and code snippets do not include specific sanitization or filtering logic for content retrieved from external integrations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 02:56 PM
Security Audit — agent-trust-hub — connect