pipefy
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from npm, which is the official command-line interface provided by the vendor.
- [COMMAND_EXECUTION]: Several commands are executed via the membrane CLI to handle authentication, search for actions, and run Pipefy operations. These are standard and necessary for the skill's stated purpose.
- [PROMPT_INJECTION]: The skill ingests data from external Pipefy resources, presenting a potential surface for indirect prompt injection. 1. Ingestion points: Data returned from action run and request commands in SKILL.md. 2. Boundary markers: Absent; no specific delimiters are defined for external content. 3. Capability inventory: Subprocess execution of the Membrane CLI. 4. Sanitization: Absent.
Audit Metadata