uipath-maestro-bpmn

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install the @uipath/cli package globally via npm or bun. This is a core dependency for the skill's operations and is a resource owned by the vendor, 'uipath'.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes and modifies existing .bpmn files and registry data. Ingesting this external content creates a potential attack surface for indirect prompt injection.
  • Ingestion points: The agent reads existing .bpmn files and JSON registry evidence from the local filesystem.
  • Boundary markers: The skill utilizes CDATA sections for script and JSON bodies within BPMN XML, though it does not define specific LLM-oriented delimiters for all processed data.
  • Capability inventory: The skill possesses the Bash tool to execute uip CLI commands and has file-writing capabilities through the Write and Edit tools.
  • Sanitization: Comprehensive guidelines are provided in the public-safety.md reference for redacting sensitive information and escaping XML characters, which serves as a mitigation strategy.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:04 PM
Security Audit — agent-trust-hub — uipath-maestro-bpmn