pipefy-introspection

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of external data from the Pipefy GraphQL API, such as card titles, metadata, and schema descriptions. This content could be manipulated by a third party to include malicious instructions designed to influence the behavior of the AI agent.\n
  • Ingestion points: The introspect_type, search_schema, and execute_graphql tools retrieve data directly from external Pipefy API endpoints.\n
  • Boundary markers: The skill does not explicitly instruct the agent to use delimiters or ignore instructions when processing strings retrieved from the API.\n
  • Capability inventory: The execute_graphql tool provides write access (mutations) to the Pipefy environment, increasing the potential impact of a successful injection.\n
  • Sanitization: There are no instructions for validating or sanitizing retrieved GraphQL data before it is processed by the agent context.\n- [COMMAND_EXECUTION]: The skill documents and enables the use of the pipefy CLI for arbitrary GraphQL execution. While mutation commands include a safety guardrail requiring the --yes flag, this capability provides the agent with extensive control over the user's Pipefy resources compared to dedicated, constrained tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 07:09 PM
Security Audit — agent-trust-hub — pipefy-introspection