
Doriku
io.dorikuPublished by doriku.ioOfficial publisherv1.2.0Updated Nov 2, 2026
Shared control plane for AI coding agents — tasks, memory, decisions, file locks. 12 tools.
Overview
Doriku is a remote MCP governance plane for AI coding agents, sharing tasks, memory, decisions and file locks across IDEs.
- What you can do
- Create, update and track engineering tasks, logging progress and results
- Share memory and decisions across agents working in parallel
- Coordinate agents with file-lock policies to prevent conflicts
- Run, monitor and cancel multi-step agent workflows
- View and set workspace-level daily token cost caps
- Generate summary snapshots of shared state for teammate review
- When to use it
- Useful for solo developers or teams running several coding agents (Claude Code, Cursor, Windsurf) in parallel who need shared task state, file locks, cost caps and audit visibility across agents.
- Requirements
- A remote endpoint at api.doriku.io over streamable HTTP. Requires a Doriku account and an API key created in the console, sent as a bearer token in the Authorization header. The companion CLI runs via npx or a global npm install and writes MCP config and Claude Code hooks.
Installation
In SourceWeft
- Open Doriku in the dashboard and add it to a workspace.
- Connect Authorization token credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.doriku%2Ftask-manager/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"doriku": {
"type": "http",
"url": "https://api.doriku.io/mcp",
"headers": {
"Authorization": "Bearer <YOUR_TOKEN>"
}
}
}
}README
@doriku/cli
The CLI for Doriku — an MCP-native governance plane for multi-IDE coding agents. Set up MCP config and Claude Code hooks, then manage workflows, file-lock policies, and cost caps from the terminal.
Quick Start
That's it. Your MCP server config and Claude Code hooks are ready.
Recommended next workflow (AI-friendly):
- Register your agent
- Create a task
- Log progress
- Update task status/result
Recommended first-value goal (15 minutes):
- Connect MCP successfully (connection test passes)
- Create one real engineering task
- Generate one summary snapshot
- Invite one teammate to review the shared state
See docs for MCP/API contracts, first-value onboarding, and recipe examples:
Installation
Setup Flow
- Sign in at doriku.io/console and create an API key
- Run the setup command
- Enter your API key when prompted — done!
Non-Interactive Mode
What It Does
Doriku also exposes AI-friendly MCP/REST contracts (structured MCP errors, machine-readable REST error fields, and a core OpenAPI spec) to make agent retries and recovery easier.
Who This Is For
- Solo developers and teams using Claude Code, Cursor, or Windsurf in parallel
- Anyone who needs governance (file locks, cost caps, audit trails) across multi-IDE agents
- Teams that need approvals, auditability, and operational visibility across agent work
CLI Reference
Workflow Subcommands
Lock Subcommands
Cost Subcommands
Links
License
MIT
Source: README.md at commit a9b3818
Tools
0Version history
1- v1.2.0LatestSep 16, 2026
Score
75% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listedRequired15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.