#7474th of 89 in Agent platforms
pilotdeck
Workspace-based agent platform with editable memory, model routing and background tasks
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.6k
- License
- AGPL-3.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- TypeScript
Missed the latest GitHub check; these facts may be older.
Overview
PilotDeck is a Node.js/TypeScript agent system that isolates files, memory and skills per WorkSpace, with a Web UI, CLI and IM front-ends. It routes simpler sub-tasks to cheaper models, exposes stored memory entries for editing and rollback, and can run tasks in the background and write results to local files. Providers are configured in ~/.pilotdeck/pilotdeck.yaml or through onboarding, and MCP is supported natively.
Who it is for: Self-hosters running several long-lived agent projects with cost control
Strengths
- Per-WorkSpace isolation of files, memory and skills limits cross-project context bleed
- Memory entries can be viewed, edited, pinned and rolled back after Dream Mode consolidation
- Smart Routing sends sub-tasks to lighter models; README reports $3.15 vs $18.36 on 7 tasks
- Supports OpenAI, Anthropic, Gemini, DeepSeek, Qwen, Kimi, MiniMax and local Ollama
Weaknesses
- AGPL-3.0 license constrains use in hosted or proprietary products
- Requires Node.js 22.13+ but below 23; native modules can need build tools on Windows
- Benchmarks and cost figures are the authors' own; PilotDeck trails Claude Code on GLM-5.3
- Install is a curl-to-bash script that clones and builds the repo locally
What it needs
- no GPU
- Docker + Compose
- Needs Node.js 22.13+ (<23), pnpm via corepack, Model provider API key or local Ollama
- Models: OpenAI-compatible, Anthropic, Google Gemini, DeepSeek, Qwen
- port 3001
Also in Agent platforms
See all 89| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |