afas-software
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the installation of the '@membranehq/cli' package via npm, which is a trusted vendor tool required for its operation.
- [COMMAND_EXECUTION]: The skill guides the agent to use standard 'membrane' CLI commands for connecting to services and running actions; these are intended functionalities for the integration.
- [DATA_EXFILTRATION]: No sensitive credential exposure was found, as the skill utilizes Membrane's managed authentication system which handles secrets server-side.
- [PROMPT_INJECTION]: Ingestion points: The skill fetches external data from AFAS Software records and tasks via the CLI. Boundary markers: None specified in the documentation. Capability inventory: The agent can execute shell commands and perform proxied network requests. Sanitization: No specific input/output sanitization processes are described.
Audit Metadata