OpenHarness
A local MCP server that discovers and shares session file paths between AI coding agents, enabling cross-referencing without modifying existing workflows.
README
OpenHarness
AI 코딩 에이전트의 최근 세션 파일을 찾아 서로 참조하게 해주는 로컬 MCP 서버.
전용 하네스 없이 하네스처럼 동작한다. 기존 도구 환경을 그대로 유지하면서, 에이전트 간 세션을 연결하는 디스커버리 레이어.
Supported Agents (MVP)
| Agent | Session Path | Format |
|---|---|---|
| Claude Code | ~/.claude/projects/ |
JSONL |
| Codex CLI | ~/.codex/sessions/ |
JSONL |
| Gemini CLI | ~/.gemini/tmp/<hash>/chats/ |
JSON |
Prerequisites
- Each agent must be able to read local files directly.
- OpenHarness only provides paths and metadata — it does not transform, relay, or store session content.
Install
npm install -g @freebird-ai/openharness
openharness init
This registers OpenHarness as an MCP server in each agent's config.
MCP Tools
get_agent_sessions
Returns recent session paths for all agents, filtered by project.
Input:
{ "project": "/Users/me/my-app" }
Output:
[
{
"agent": "claude-code",
"path": "/Users/me/.claude/projects/-Users-me-my-app/session.jsonl",
"format": "jsonl",
"updatedAt": "2026-03-17T14:32:00Z",
"matchedProject": "/Users/me/my-app",
"matchedBy": "directory-name"
}
]
get_agent_session
Returns the recent session path for a specific agent.
Input:
{ "agent": "claude-code", "project": "/Users/me/my-app" }
Output:
{
"agent": "claude-code",
"path": "/Users/me/.claude/projects/-Users-me-my-app/session.jsonl",
"format": "jsonl",
"updatedAt": "2026-03-17T14:32:00Z",
"matchedProject": "/Users/me/my-app",
"matchedBy": "directory-name"
}
Returns null if no session is found.
Project Matching
| Agent | Strategy |
|---|---|
| Claude Code | Project path escaped with - as directory name |
| Codex CLI | Workspace field inside session JSONL |
| Gemini CLI | Project path hash matching tmp/<hash>/ |
How It Works
User → Codex: "Review what Claude Code worked on recently"
Codex → OpenHarness MCP: get_agent_session({ agent: "claude-code", project: "..." })
Codex ← receives session path + metadata
Codex → reads the JSONL file directly and performs review
No rules needed in agent instruction files. MCP tool descriptions are enough for agents to use naturally.
Manual MCP Configuration
If you prefer to configure manually instead of using init:
{
"mcpServers": {
"openharness": {
"command": "openharness",
"args": []
}
}
}
Add this to your agent's MCP config file:
| Agent | Config File |
|---|---|
| Claude Code | ~/.claude/mcp.json or project .mcp.json |
| Codex CLI | project .codex/ |
| Gemini CLI | ~/.gemini/settings.json |
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.