connect
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches an installation script from the vendor domain at https://composio.dev/install.
- [REMOTE_CODE_EXECUTION]: Executes the downloaded setup script using a pipe-to-shell operation (curl -fsSL https://composio.dev/install | bash).
- [COMMAND_EXECUTION]: Employs the 'composio' CLI to interact with external services and allows for executing custom workflows via the 'composio run' command.
- [DATA_EXFILTRATION]: Provides features to send data to third-party APIs (Gmail, Slack, GitHub, etc.) and arbitrary URLs using the 'composio execute' and 'composio proxy' tools.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. * Ingestion points: Input parameters for Gmail and Slack tools in SKILL.md. * Boundary markers: Not present. * Capability inventory: Command execution via 'composio' CLI and network operations. * Sanitization: No sanitization or validation logic is specified in the skill body.
Audit Metadata