proletariat-cli MCP server
✓ VerifiedAn agent orchestration platform that allows AI clients to spawn, coordinate, and manage multi-agent development workflows using tickets and isolated workspaces.
Also known as chrismcdermut/proletariat, @proletariat/cli
- Stars
- 36
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 0.3.134
- Last commit
- Jun 9, 2026
- License
- Apache-2.0
What is proletariat-cli MCP?
proletariat-cli is an agent orchestration platform that allows AI clients to spawn, coordinate, and manage multi-agent development workflows using tickets and isolated workspaces.
Summary generated from the project's README
About proletariat-cli MCP server
Managing multi-agent development workflows through structured tickets and isolated workspaces is what proletariat-cli enables for developers. This orchestration platform allows AI clients to spawn, coordinate, and manage workers that operate within secure Docker containers. By using a command-line interface, users can create tickets for specific features, spawn agents to implement them, and have those agents automatically open pull requests once the code is written.
Users can set up the platform locally via Homebrew or npm. The system provides durability through persistent Tmux sessions and prevents code conflicts by assigning each agent its own git branch and worktree. For those using AI-native environments, the built-in MCP server provides over 100 tools that can be integrated into clients like Claude Code or Cursor. The platform supports different agent naming themes, such as billionaires or Toyota models, and offers a JSON mode for programmatic control by other AI agents.
What you can do with it
- 1Automate feature implementationAn AI client can create a feature ticket and spawn an agent to write the code and open a pull request.
- 2Maintain secure agent environmentsDevelopers can run AI workers in sandboxed containers to ensure they do not affect the host machine.
- 3Coordinate complex multi-agent tasksUsers can manage a structured workflow where tickets accumulate context and allow for hand-offs between different agents.
- 4Programmatic agent orchestrationAI agents can drive the CLI directly using machine-readable output to manage work.
Written from the project's README. Every use case is checked against a quote from it.
How to install proletariat-cli MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall proletariat-cli MCP in Claude Code
$ claude mcp add proletariat-cli -- npx -y @proletariat/cliInstall proletariat-cli MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"proletariat-cli": {
"command": "npx",
"args": [
"-y",
"@proletariat/cli"
]
}
}
}Install proletariat-cli MCP in VS Code
.vscode/mcp.json
{
"servers": {
"proletariat-cli": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@proletariat/cli"
]
}
}
}Install proletariat-cli MCP in VS Code CLI
$ code --add-mcp '{"name":"proletariat-cli","type":"stdio","command":"npx","args":["-y","@proletariat/cli"]}'Install proletariat-cli MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"proletariat-cli": {
"command": "npx",
"args": [
"-y",
"@proletariat/cli"
]
}
}
}Install proletariat-cli MCP in Codex
$ codex mcp add proletariat-cli -- npx -y @proletariat/cliInstall proletariat-cli MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.proletariat-cli] command = "npx" args = ["-y", "@proletariat/cli"]
Install proletariat-cli MCP in Gemini CLI
$ gemini mcp add -s user proletariat-cli npx -- -y @proletariat/cliInstall proletariat-cli MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"proletariat-cli": {
"command": "npx",
"args": [
"-y",
"@proletariat/cli"
]
}
}
}Automated checks
Ran Nov 14, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablechrismcdermut/proletariat is public and reachablegithub.com/chrismcdermut/proletariat ↗Pass
- Package names this repository as its source@proletariat/cli names chrismcdermut/proletariat as sourcenpmjs.com/package/@proletariat/cli ↗Pass
- Build provenance (Sigstore attestation)Sigstore attestation for @proletariat/clinpmjs.com/package/@proletariat/cli/v/0.3.134#provenance ↗Pass
- Open source license (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-06-09github.com/chrismcdermut/proletariat ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from @proletariat/cliPass
- No critical advisory in the package or its dependencies400 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHub██████╗ ██████╗ ██████╗ ██╗ ███████╗████████╗ █████╗ ██████╗ ██╗ █████╗ ████████╗
██╔══██╗██╔══██╗██╔═══██╗██║ ██╔════╝╚══██╔══╝██╔══██╗██╔══██╗██║██╔══██╗╚══██╔══╝
██████╔╝██████╔╝██║ ██║██║ █████╗ ██║ ███████║██████╔╝██║███████║ ██║
██╔═══╝ ██╔══██╗██║ ██║██║ ██╔══╝ ██║ ██╔══██║██╔══██╗██║██╔══██║ ██║
██║ ██║ ██║╚██████╔╝███████╗███████╗ ██║ ██║ ██║██║ ██║██║██║ ██║ ██║
╚═╝ ╚═╝ ╚═╝ ╚═════╝ ╚══════╝╚══════╝ ╚═╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚═╝╚═╝ ╚═╝ ╚═╝
Seize the means of production - Ship 100x.
Agent orchestration platform for AI labor. Spin up workers for all work, on demand.
Themed agents, including billionaires - Finally, they work for us.
⚠️ Beta Software — Under active development. Commands and APIs may change between versions, and bugs are actively being squashed.
Let's get you shipping [Book a call][cal-link] - I'm happy to help you get prlt running or chat feedback, ideas, multi-agent workflows, and the future of work/labor (and economic labor theory..)
TLDR
prlt is an agent orchestration platform for AI labor. Spin up workers on demand, coordinate multi-agent development from one CLI. Isolated workspaces, secure containers, persistent state.
brew install chrismcdermut/proletariat/prlt # macOS (Homebrew)
# or
npm install -g @proletariat/cli # any platform (npm)
prlt new
prlt ticket create --title "Add OAuth" --category feature
prlt work spawn # Interactive: select tickets, environment, action
Agent spawns in its own branch, writes code, opens PR. You review and merge.
Why prlt?
- Isolated - Each agent gets its own git branch. No conflicts.
- Secure - Docker containers, sandboxed from your host.
- Durable - Tmux sessions persist. Close window, agent keeps working.
- Trackable - One database, one CLI, all your agents.
- Ephemeral - Spawn agents on demand. They work, they PR, they're done.
- Structured - Tickets provide structured context, not freeform chat.
- Persistent - Tickets accumulate context over time. Hand off between agents.
- Agent-native -
--jsonmode lets AI agents drive the CLI programmatically.
Quick Start
brew install chrismcdermut/proletariat/prlt # Install (Homebrew, recommended)
# or
npm install -g @proletariat/cli # Install (npm, all platforms)
prlt new # Create HQ, add repos, choose theme
prlt ticket create --title "Add OAuth" --category feature
prlt work spawn # Interactive: select tickets, environment, action
# Agent creates PR → You review → Merge → Done
sequenceDiagram
participant You
participant prlt
participant Agent
participant GitHub
You->>prlt: prlt ticket create
You->>prlt: prlt work spawn
prlt->>prlt: Create branch
prlt->>prlt: Create workspace
prlt->>Agent: Spawn agent
Agent->>Agent: Read ticket
Agent->>Agent: Write code
Agent->>GitHub: Commit
Agent->>GitHub: Open PR
Agent->>prlt: Update status
prlt->>You: PR ready
You->>GitHub: Review & approve
Spawn agents to implement, groom, or review—not just write code.
Interactive Menus
prlt work guides you through project and ticket selection:
Choose your operation—start a single agent, batch spawn, or watch a column:
Select tickets to spawn, grouped by priority:
Deep Dive
| Problem | Solution |
|---|---|
| Agents conflict with each other's changes | Isolated - Each agent gets its own git branch and worktree |
| Agents run unsandboxed on your machine | Secure - Docker containers, sandboxed from your host (looking into host sandbox options) |
| You lose track of who's doing what | Trackable - All state in one SQLite database, one CLI |
| Sessions die when you close a window | Durable - Tmux sessions persist, detach/reattach anytime |
| Context scattered across chat windows | Structured - Tickets with requirements, acceptance criteria |
| Starting agents is heavyweight | Ephemeral - Spawn on demand, they work, they PR, they're done |
| Context lost between agent runs | Persistent - Tickets accumulate context, hand off between agents |
Installation
Homebrew (recommended)
brew install chrismcdermut/proletariat/prlt
Works on both Apple Silicon (arm64) and Intel (x86_64) Macs. No compiler needed.
Upgrade:
brew update
brew upgrade prlt
npm / pnpm (all platforms)
npm install -g @proletariat/cli
# or
pnpm install -g @proletariat/cli
pnpm 10+ note: pnpm 10 blocks native addon build scripts by default. If
prltfails with a native module error after install, runpnpm approve-buildsin the global store and reinstall, or usenpm/brewinstead.
Verify:
prlt --version
MCP Server
prlt includes a built-in MCP server with 100+ tools. Add it to your AI client:
Claude Code (~/.claude.json):
{
"mcpServers": {
"prlt": { "command": "prlt", "args": ["mcp-server"] }
}
}
Cursor / Other clients (via npx):
{
"mcpServers": {
"prlt": { "command": "npx", "args": ["-y", "@proletariat/cli", "mcp-server"] }
}
}
Listed on: MCP Registry | npm
Data Model
Workspace (HQ)
├── Projects
│ ├── Epics → Tickets
│ └── (references a Workflow)
├── Workflows → Phases → Statuses (can be shared across projects)
├── Specs (can span projects)
├── Actions (reusable templates)
├── Agents
│ ├── Staff (persistent, named)
│ └── Temp (ephemeral, per-ticket)
└── Executions (running sessions)
├── Docker
│ ├── Tmux session
│ ├── Terminal or Background display
│ └── Safe or YOLO permissions
└── Host
├── Tmux session
├── Terminal or Background display
└── Safe or YOLO permissions
| Entity | Description |
|---|---|
| Project | Groups tickets and epics, references a workflow |
| Epic | Work container with lifecycle (draft → active → complete) |
| Ticket | Individual work item with requirements and acceptance criteria |
| Spec | Static documentation (can span projects, linked to epics) |
| Workflow | Status flow configuration (can be shared across projects) |
| Phase | Stage in a workflow |
| Status | Ticket state within a phase |
| Action | Reusable prompt/action templates |
| Agent (Staff) | Persistent named agent with dedicated workspace |
| Agent (Temp) | Ephemeral agent spawned for a single ticket |
| Execution | Running agent session on a ticket |
| Display | Terminal (new tab) or Background (detached) |
Example Workflow
A workflow defines how tickets move through your process. Projects reference a workflow, and multiple projects can share the same one.
Kanban Workflow
├── Backlog # New tickets land here
├── In Progress # Agent working (prlt work spawn)
├── Review # PR ready (prlt work ready)
└── Done # Merged (prlt work complete)
Scrum Workflow
├── Backlog
├── Sprint
│ ├── To Do
│ ├── In Progress
│ └── In Review
└── Done
Tickets flow through statuses as work progresses. Agents automatically move tickets when they start work, open PRs, or complete tasks.
Workspace Structure
Each agent gets a copy of all repos (repo scoping coming soon). Work happens on isolated branches.
my-project/
├── .proletariat/
│ └── workspace.db # Tickets, executions, state
├── repos/
│ ├── frontend/ # Your repos
│ ├── backend/
│ └── infra/
└── agents/
├── staff/
│ └── alice/ # Named agent with persistent workspace
│ ├── frontend/
│ ├── backend/
│ └── infra/
└── temp/
├── agent-abc123/ # Ephemeral: Working on TKT-042 (OAuth)
│ ├── frontend/ # All repos on branch: feat/TKT-042-oauth
│ ├── backend/
│ └── infra/
└── agent-def456/ # Ephemeral: Working on TKT-043 (API)
├── frontend/ # All repos on branch: feat/TKT-043-api
├── backend/
└── infra/
Agent Naming Themes
Themes control how agents are named. Staff agents use theme names directly (e.g., bezos, camry). Ephemeral agents add an adjective prefix (e.g., bold-bezos, keen-camry). Currently ephemeral names also include a number suffix (bold-bezos-1), but this will be removed soon.
Built-in Themes:
| Theme | Description | Example Names |
|---|---|---|
billionaires |
Tech founders & executives (default) | musk, gates, bezos |
toyotas |
Toyota vehicle models | camry, supra, tacoma |
companies |
Major tech companies | stripe, vercel, linear |
billionaires — Finally, they work for us.
Theme Commands:
prlt agent themes list # List available themes
prlt agent themes set billionaires # Set active theme
prlt agent themes create mytheme # Create custom theme
prlt agent themes add-names mytheme # Add names to custom theme
Themes are selected during prlt new.
Three Ways to Use Commands
1. Interactive (Humans)
Run without flags—get guided prompts:
$ prlt ticket create
? Title: Add password reset
? Description: Email-based password reset flow
? Priority: P1
? Category: feature
✓ Created TKT-043
View ticket details with prlt ticket:
2. JSON Mode (AI Agents)
Add --json for machine-readable output:
$ prlt work start --json
{
"prompt": {
"type": "list",
"message": "Select ticket to work on:",
"choices": [
{
"name": "[P1] TKT-042 - Add user authentication",
"value": "TKT-042",
"command": "prlt work start TKT-042 --json"
}
]
}
}
AI agents parse this, make selections, call the next command.
3. Flags (Scripts/CI)
Pass everything directly:
prlt ticket create \
--title "Add OAuth" \
--description "Google and GitHub OAuth" \
--priority P1 \
--category feature
Execution Modes
Environment - where the agent runs:
| Environment | Flag | Best For |
|---|---|---|
| 🐳 Docker | (default if devcontainer exists) | Safety—fully isolated container |
| 🏃 Host | --run-on-host |
Speed—no container overhead |
Display - how you see it:
| Display | Flag | Best For |
|---|---|---|
| 📺 Terminal | --display terminal |
Watch in new terminal tab |
| 🔇 Background | --display background |
Detached, reattach later |
Permissions - agent access level:
| Mode | Flag | Description |
|---|---|---|
| 🔒 Safe | (default) | Agent prompts for permissions |
| 🕺 YOLO | --skip-permissions |
No prompts, full access. Use with Docker for safe autonomy. |
All sessions run in tmux under the hood—close the window, agent keeps working.
# Default: Docker + terminal (if devcontainer exists)
prlt work start TKT-042
# Docker + background
prlt work start TKT-042 --display background
# Host + background (fast, no container)
prlt work start TKT-042 --run-on-host --display background
# Docker + YOLO (full autonomy, safely sandboxed)
prlt work start TKT-042 --skip-permissions
Parallel Agents
Work on multiple tickets simultaneously.
Interactive (humans):
$ prlt work spawn
? Spawn mode: Select specific tickets
? Select tickets:
◉ [P1] TKT-042 - Add user authentication
◉ [P1] TKT-043 - Add API rate limiting
◯ [P2] TKT-044 - Add email notifications
? Action: implement
? Environment: docker
Spawning 2 tickets...
JSON mode (AI agents): (multi-select WIP)
$ prlt work spawn --json --many
{
"prompt": {
"type": "checkbox",
"message": "Select tickets to spawn:",
"choices": [
{"name": "[P1] TKT-042 - Add user authentication", "value": "TKT-042"},
{"name": "[P1] TKT-043 - Add API rate limiting", "value": "TKT-043"}
]
}
}
Flags (scripts/CI):
prlt work spawn TKT-042 TKT-043 --action implement --mode docker
Each agent works in its own branch. No conflicts.
Scaling: The main limit is your machine. 50+ concurrent agents is achievable—depends on CPU, RAM, and whether you're running Docker or host mode.
Monitor running agents with prlt execution:
flowchart LR
subgraph You
spawn[prlt work spawn]
end
subgraph Agents
A1[Agent 1<br/>TKT-042 OAuth]
A2[Agent 2<br/>TKT-043 Rate Limit]
A3[Agent 3<br/>TKT-044 Notifications]
end
subgraph GitHub
PR1[PR #101<br/>feat/TKT-042-oauth]
PR2[PR #102<br/>feat/TKT-043-rate-limit]
PR3[PR #103<br/>feat/TKT-044-notifications]
end
spawn --> A1
spawn --> A2
spawn --> A3
A1 --> PR1
A2 --> PR2
A3 --> PR3
Agent-created PRs ready for review:
Command Reference
Run prlt <command> --help for flags and options.
Use Cases
Parallel Feature Development
proletariat-cli MCP questions
How do I add proletariat-cli MCP to Claude Code?
Run claude mcp add proletariat-cli -- npx -y @proletariat/cli in your terminal. Add --scope user to make it available in every project.
How do I add proletariat-cli MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How does the platform ensure agents don't conflict with each other?
Each agent is isolated within its own git branch and worktree.
Is the execution environment secure?
Yes, agents run in Docker containers, which sandboxes them from your host machine.
Can I use this with Claude Code or Cursor?
Yes, it includes a built-in MCP server that can be added to Claude Code via ~/.claude.json or to Cursor and other clients via npx.
What happens if I close my terminal window while an agent is working?
The sessions are durable because Tmux sessions persist, allowing you to detach and reattach anytime.