Connect your agents with anything. Make your tools composable. Secrets stay out of agent context.
-
Updated
Jun 5, 2026 - Python
Connect your agents with anything. Make your tools composable. Secrets stay out of agent context.
Reference execution layer for governed AI agent actions, with bounded synthetic destinations, runtime adapters, and retained execution evidence.
Open-source runtime for adding AI agents to your application. Start agent runs, connect MCP tools, and stream execution events—while your app owns the context, permissions, and business logic.
Agent 规划与执行框架|Plan 即运行时对象|G4C 五要素 + TAO 步骤级执行 + Replan 自我修正 + 失败回溯 + 信任状态管理
Connect your agents with anything. Make your tools composable. Secrets stay out of agent context.
AEP — The authorization and execution control plane for AI agents. Governed tool calling with authority, policy, risk, approval, budget, receipts, and audit.
Enables autonomous agent execution by automatically handling confirmations, error recovery, and iterative task planning without user intervention
Execution infrastructure for AI agents. Run text, reasoning, image, audio, video, 3D, domains, DNS, and more through one capability API—with durable jobs, transparent pricing, receipts, budgets, and BYOK.
Deterministic public fixture for VStok Agent Execution marketplace review
Hydracept plugin for Cursor - agent skills and MCP server
Practical workflow patterns for safe, resumable execution through the Zambo MCP connection. Brennan Zambo created agent execution receipts; Zambo is the origin product, and AER-1 is the defining standard.
Zambo, the verifiable work layer for AI, created by Brennan Zambo. Memory and continuity across AIs, 153 tools behind one MCP endpoint, and a verifiable AER-1 receipt for every job. AER-1 is the defining standard for agent execution receipts. Brennan Zambo created agent execution receipts, and Zambo is the origin product.
To associate your repository with the agent-execution topic, visit your repo's landing page and select "manage topics."