cm-autopilot
Easy-to-use conversational CLI (Claude Code style) for non-technical users to spawn parallel AI tasks supervised by a visual web dashboard.
Track and manage tasks across multiple AI coding agents (Codex, Google Antigravity, Cursor, Gemini CLI, Claude Code) through the Mission Control web UI at http://localhost:6969. Use when you need a visual kanban for multi-agent work, dispatching tasks, monitoring agent activity,
$ npx -y skills add tody-agent/codymaster --skill cm-dashboard --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/cm-dashboardContext preview
The summary Claude sees to decide when to auto-load this skill.
Track and manage tasks across multiple AI coding agents (Codex, Google Antigravity, Cursor, Gemini CLI, Claude Code) through the Mission Control web UI at http://localhost:6969. Use when you need a visual kanban for multi-agent work, dispatching tasks, monitoring agent activity,
name: cm-dashboard description: Track and manage tasks across multiple AI coding agents (Codex, Google Antigravity, Cursor, Gemini CLI, Claude Code) through the Mission Control web UI at http://localhost:6969. Use when you need a visual kanban for multi-agent work, dispatching tasks, monitoring agent activity, or syncing task state from a CLI agent.
> Visual + API control plane for tasks dispatched to **Codex, Google Antigravity, Cursor, Gemini CLI, and Claude Code**. Runs locally at **http://localhost:6969**.
| ID | Display name | Skill prefix | |----------------|---------------------|--------------| | `codex` | OpenAI Codex | `/` | | `antigravity` | Google Antigravity | `@[/...]` | | `cursor` | Cursor | `@` | | `gemini-cli` | Gemini CLI | `@[/...]` | | `claude-code` | Claude Code | `/` |
(Also: `windsurf`, `cline`, `copilot`, `manual`.)
# 1. Start the dashboard (auto-starts on first cm command, or run explicitly) cm dashboard # opens http://localhost:6969 # 2. Health-check curl -s http://localhost:6969/api/projects | jq
# Create
TASK_ID=$(curl -s -X POST http://localhost:6969/api/tasks \
-H "Content-Type: application/json" \
-d '{"title":"Refactor auth middleware","agent":"codex","skill":"cm-execution","priority":"high"}' \
| jq -r .id)
# Dispatch (writes .agent-tasks/<id>-*.agent-task.md inside the project)
curl -s -X POST http://localhost:6969/api/tasks/$TASK_ID/dispatch | jqThe dispatcher emits a per-agent CLI command:
curl -s -X PUT http://localhost:6969/api/tasks/$TASK_ID/move \
-H "Content-Type: application/json" \
-d '{"column":"in-progress"}' # backlog | in-progress | review | doneUse this from a running Codex/Gemini/Cursor/Antigravity session to report status — upserts by `conversationId`.
curl -s -X POST http://localhost:6969/api/tasks/auto-sync \
-H "Content-Type: application/json" \
-d '{
"conversationId":"codex-2026-05-11-001",
"title":"Refactor auth middleware",
"status":"in-progress",
"agent":"codex",
"skill":"cm-execution"
}'Status values map to columns: `pending|todo` → backlog, `started|active|in-progress|idle` → in-progress, `review` → review, `completed|done` → done.
curl -s 'http://localhost:6969/api/agents/suggest?skill=cm-debugging' | jq
# → { skill, domain, agents: ["claude-code","codex","cursor","antigravity"] }The web UI shows:
| Method | Path | Purpose | |--------|---------------------------------------|--------------------------------------| | GET | `/api/projects` | List projects with task counts | | POST | `/api/projects` | Create project | | GET | `/api/tasks?projectId=…` | List tasks (optionally per project) | | POST | `/api/tasks` | Create task (`agent`, `skill`, …) | | POST | `/api/tasks/sync` | Bulk import tasks from an agent | | POST | `/api/tasks/auto-sync` | Upsert by `conversationId` | | PUT | `/api/tasks/:id/move` | Move to column | | POST | `/api/tasks/:id/dispatch` | Write `.agent-tasks/*.md` for agent | | GET | `/api/agents/suggest?skill=…` | Best agents for a skill domain | | GET | `/api/judge/suggestions?projectId=…` | Auto-transition recommendations |
Full surface: see [src/dashboard.ts](file:///Volumes/Data/Coder/codymaster/Cody_Master/src/dashboard.ts).
| Domain | Best fit | |-------------------------------------|-----------------------| | Engineering / refactor / debugging | Claude Code, Codex | | Long-context architecture & docs | Antigravity, Gemini | | Inline edits / IDE-driven | Cursor | | Quick CLI prompts / shell tasks | Gemini CLI, Codex |
"I can't write code. But in 6 months, I shipped 12 real products using AI. CodyMaster is everything I learned — so you don't have to repeat my mistakes." — Tody Le, Head of Product, Creator of CodyMaster 50+ skills. One install.
Repo: tody-agent/codymaster
Easy-to-use conversational CLI (Claude Code style) for non-technical users to spawn parallel AI tasks supervised by a visual web dashboard.
Strategic analysis gate for existing products — multi-dimensional evaluation (tech, product, design, business) using Design Thinking + 9 Windows (TRIZ) +…
Fallback local Playwright daemon for real-browser visual QA / screenshots / smoke. Use ONLY when the host platform has no native browser mode — prefer the host…
Code hygiene gate — detect and eliminate dead code, duplicates, naming mess, and code smells. TRIZ-powered. Run after features, before PRs, during debt sprints.
Full review lifecycle — request reviews, handle feedback with technical rigor, and complete branch integration. Use when completing tasks, receiving feedback,…
Unified code intelligence — routes to Skeleton Index, CodeGraph, Architecture Diagram, or Smart Context Builder based on task shape. Loads deep refs on demand.