agentteams-hiclaw-multi-agent-orchestration
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides installation commands that download and execute scripts from the official Higress project domains (
higress.aiandhigress.io) using Bash and PowerShell. - [COMMAND_EXECUTION]: Configuration examples for Model Context Protocol (MCP) servers involve executing local shell commands, such as
npxandpython, to facilitate agent interactions with the system. - [PROMPT_INJECTION]: The orchestration platform processes data from external sources, presenting an inherent surface for indirect prompt injection.
- Ingestion points: Matrix chat room messages and tool outputs from MCP servers (SKILL.md).
- Boundary markers: Not explicitly defined in the provided configuration templates.
- Capability inventory: Access to the local file system (
/workspace) and shell command execution viastdioMCP (SKILL.md). - Sanitization: No specific input sanitization or filtering logic is detailed in the resource definitions.
- [SAFE]: Sensitive information, including API keys and administrative passwords, is managed through environment variables and Kubernetes secrets, avoiding unsafe hardcoded credentials.
Audit Metadata