Agenthold

Name: Agenthold
Author: edobusy

by edobusycommunityv0.5.1

agenthold is an MCP server for coordinating shared, versioned state between agents with optimistic concurrency control and conflict detection. It provides claim/release/status/wait coordination tools so multiple agents can safely work on the same resources without silently overwriting each other.

From the registry: Shared versioned state for multi-agent AI workflows

GitHub Repository Website

Install & configure

Prerequisites

required
Install agenthold
Install the Python package from PyPI. The README shows both pip and uv pip options.
```
$ pip install agenthold
```
https://pypi.org/project/agenthold/

Install in your client

Paste into the AI you're chatting with — it will figure out the rest

Please install the `agenthold` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install agenthold** — Install the Python package from PyPI. The README shows both pip and uv pip options. Run: `pip install agenthold` (https://pypi.org/project/agenthold/)

Canonical MCP server config (stdio transport):
- command: `agenthold`
- args: ["--db","<path-to-state.db>"]

Note: Transport shown in the README is stdio. The database file is created automatically on first run. Optional flags include --tools advanced, --claim-ttl <seconds>, and repeatable --workspace <name=path>.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.

Note: Transport shown in the README is stdio. The database file is created automatically on first run. Optional flags include --tools advanced, --claim-ttl <seconds>, and repeatable --workspace <name=path>.