coordinating-multi-space-devops-agent

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [Multi-Space Orchestration]: The skill coordinates actions across multiple AWS environments (e.g., prod, staging, knowledge). It uses authenticated tools (aws_devops_agent__list_agent_spaces, aws_devops_agent__chat) to route queries and synthesize information from different accounts.
  • [Local Configuration Management]: The skill reads from and offers to write to local configuration files (e.g., .claude/aws-agents-for-devsecops.md or AGENTS.md). This is used to maintain a persistent routing table for the session, which helps the agent identify which space corresponds to specific environments.
  • [Data Ingestion and Forwarding]: A potential security consideration involves Pattern B, where data retrieved from one 'knowledge' AgentSpace is forwarded as input into an 'investigation' in another space. While this is a functional requirement for bridging state between isolated environments, it represents an indirect data flow where responses from one remote source influence actions in another environment.
  • [Authentication Requirements]: The skill explicitly notes the requirement for SigV4 authentication for multi-space routing, as standard tokens are typically scoped to a single environment. This ensures that the underlying MCP server handles identity and access management appropriately across account boundaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 07:17 AM
Security Audit — agent-trust-hub — coordinating-multi-space-devops-agent