agent-supervision

Temporary rooms for Humans and AI Agents — run, supervise and steer Agent Tasks without a permanent workspace.

An agent supervisor and software factory foreman powered by decision models

mutual layer for multiple agents from different environment or providers to work on same interface or device, toward the same final goal

Open-source agent execution-recovery controller for coding and tool-using AI agents.

ChatGPT <-> Codex bridge: local coding agents ask ChatGPT for advice, and ChatGPT can inspect or steer your running Codex sessions. No API key.

Open-source, self-hosted browser workspace for running and supervising Codex, Claude Code, OpenCode, Pi, and other AI coding agents.

DSH task supervision with DAG progress, plan review, persistent control and stage/completion checks.

Delegate coding tasks to agents, collect artifacts, and verify results locally with standalone supervision skills for Claude, Codex, and Kimi automation.

Send Pi agents back to work when they stop before the job is done.

AI coding that actually makes you move with superhuman speed.

Windows-first local agent supervision skills for Claude Code, Codex, and Kimi Code. 本地 Agent 委托审查与可追溯监督。

Live supervision for OpenAI Codex agent sessions in git repositories -- worktree observation, base-ancestry checks, per-repo invariant rules. A product of Divergent Health, Inc., released under the MIT License.

DSH DevKit: multimodal gallery + multi-agent supervision console (DeepSeek Harness dynamic Cordis plugin)

MCP server for Cursor Cloud Agents (background agents): launch, follow up, supervise and track the cost of Cursor agents from Claude Code, Codex CLI or OpenCode.

Supervise an AI coding agent (Claude Code / Codex / ACP) overnight and keep control: block forbidden tool calls before they run, gate 'done' on real tests, checkpoint work to git, cap spend, and get a morning report where every line is sourced. Local-first, no telemetry. Rust daemon + Tauri app + Claude Code plugin.

User-intent tracking and agent supervision for DeepSeek Harness (DSH) · 用户意图监督与受控纠偏

Dont-Stop 插件用于保证智能体长时间持续工作：设定一个截止时间，若 AI 提前完成任务，插件将自动督促智能体继续迭代，自主推进下一步任务。

The control tower for your coding agents — one resident mind above Claude Code, Codex, opencode, zcode: every session, the machine's vitals, and the disk's actual record in one attention state. Verdicts SILENT·FLAG·ASK·DENY — never allow. Watch→contest→advise→drive. C/C++20, local spine, hash-chained ledger, 9B on your RTX. FUSOR-1. MIT.

Calibrated yes-or-no checks for AI coding agents: a Claude Code plugin and a Rust crate on the Jev decision model