agent-operations

Self-hosted control plane for AI agents: dispatch tasks, review runs, track spend, and operate OpenClaw, Claude Code, Codex, and other runtimes.

Public, source-cited Fable 5, Mythos, Claude Code, and agent-operations second brain. Not affiliated with Anthropic.

A Telegram control plane for local AI workflows with durable sessions, scheduling, recovery, and permission-aware automation.

AI-assisted setup kit for testing a low-cost open-model orchestrator/executor stack.

给AI当老板 — 21章AI Agent团队搭建实战手册 / Managing AI Agents: The Complete Playbook (Bilingual)

🖥️ SATO OS — Onchain Agent Mission Control. Build and run your onchain agent on any chain, using any model.

RCOS operator control plane for DeepSeek Harness — GoalRunner, authority gates with explicit approval, three-truth verification (execution / capability / objective satisfaction), trust ladder and next-action, task identity rebuilt from Archon runs. Side-loaded, no fork.

Astromesh Leia — Claude Code plugin for agent operations in plain English: create, deploy and monitor WhatsApp and multi-channel agents on Astromesh Nexus without touching Kubernetes.

Free starter kit for running AI coding agents with proof gates, status badges, and lane discipline. MIT.

Sample-first AI secretary proof for mailbox triage, draft-only replies, and human confirmation queues.

Local operational dashboard and guarded admin cockpit for OpenClaw

Claude Code / AIエージェントの運用統制ハンドブック — field-tested governance patterns for AI agents: work packets, independent replay, multiplicity ledgers, kill logs, fail-closed gates

Practical baseline for running AI agents on-prem. Checklists, safe defaults, and operator field notes for local-first deployments.

Decision gate that stops AI agents from burning tokens on subagents that can't deliver

AI agent operations control plane for agent registry, tool governance, knowledge tracking, run monitoring, and audit evidence.

A written doctrine for running a fleet of AI agents under real operational discipline: budgets, verification gates, contract templates, and a failure ledger.

Rules don't hold AI agents. Machines do. An open protocol for multi-agent teams, grown from real failures with numbers before and after.

Portable agent operations kit for durable, verifiable work across coding-agent harnesses.