ITHZ MCP
Local-first deterministic project memory for AI coding agents, with context packs, decisions, gates, risks, scoped claims and explicit checkpoints in project-owned files.
README
ITHZ MCP
<!-- mcp-name: dev.ithz/ithz-mcp -->
Local-first deterministic project memory for AI coding agents and MCP-compatible development workflows.
ITHZ MCP stores durable agent work memory in project-owned files: context packs, decisions, gates, risks, reviewer notes, scoped claims and explicit checkpoints. Git remains the code history and transport; ITHZ MCP keeps the operational memory that helps agents resume work without depending only on hidden chat history or repeated full-repo scans.
Public Links
- Product page: https://ithz.dev/mcp/
- PyPI package: https://pypi.org/project/ithz-mcp/
- Official MCP Registry name:
dev.ithz/ithz-mcp - Official MCP Registry API: https://registry.modelcontextprotocol.io/v0.1/servers/dev.ithz%2Fithz-mcp/versions/latest
- Package repository: https://github.com/thegobi/ithz_mcp
Install From PyPI
python -m pip install ithz-mcp
python -m ithz_mcp version
Minimal stdio MCP configuration:
{
"mcpServers": {
"ithz_mcp": {
"command": "python",
"args": [
"-m",
"ithz_mcp",
"mcp-server",
"--project",
"/path/to/project",
"--mode",
"read-only",
"--protocol",
"direct",
"--storage-profile",
"native-archive"
]
}
}
}
Use a write-enabled profile only when your host and workflow explicitly allow
the agent to append sanitized task checkpoints into project.ithz.
What It Is
ITHZ MCP is a local project-memory layer for AI-assisted development. It is designed for agents and developers who want deterministic, auditable handoff state next to the project:
- context packs for onboarding an agent to a repo or task
- durable decisions, gates, risks and next steps
- reviewer notes and scoped claim ledgers
- end-of-task checkpoints for future agents
- local-first operation through project-owned memory files
- MCP stdio integration for compatible coding-agent hosts
What It Is Not
ITHZ MCP does not replace Git, a production database, cloud sync, a source-code backup, a vector database, or every retrieval system. It complements these tools by keeping durable project memory explicit and inspectable.
Distribution Artifacts
This repository carries public package artifacts and metadata, not the full development source tree.
Direct artifact URLs:
https://github.com/thegobi/ithz_mcp/raw/main/ithz_mcp.zip
https://github.com/thegobi/ithz_mcp/raw/main/ithz-mcp-macos.zip
https://github.com/thegobi/ithz_mcp/raw/main/ithz-mcp-ubuntu.tar.gz
https://github.com/thegobi/ithz_mcp/raw/main/ithz-mcp-ubuntu.deb
https://github.com/thegobi/ithz_mcp/raw/main/ithz-native-macos-intel-build-kit.zip
Metadata:
server.json
VERSION.json
SHA256SUMS.txt
RELEASE_NOTES.md
Agent Bootstrap
For agent-assisted installation into another project:
- Copy
ithz_mcp.mdorinstall_ithz.mdinto the target project. - Ask the local coding agent to read that file and install ITHZ MCP.
- The agent should preserve an existing
project.ithzif present, configure local MCP/Git support, and report exactly what changed.
Platform Notes
macOS Intel Native Runtime
ithz-mcp-macos.zip installs the Python stdio MCP layer. Native project.ithz
storage on Mac Intel additionally needs ithz-native, which can be built from
ithz-native-macos-intel-build-kit.zip:
unzip ithz-native-macos-intel-build-kit.zip
cd ithz-native-macos-intel-build-kit
chmod +x native_ithz/macos_native/build_macos_native_package.sh
native_ithz/macos_native/build_macos_native_package.sh --arch x86_64
Then install the generated native package:
cd native_ithz/dist/macos_native/ithz-native-preview-v0.1-alpha-rc2-macos-x86_64
./smoke_test.sh
./install_macos_native.sh
export ITHZ_NATIVE_EXE="$HOME/.local/share/ithz-mcp/native/ithz-native"
Ubuntu/Linux
The Ubuntu package includes install_ithz.md, Linux host installers, user-local
wrapper scripts, a Linux native source build kit and extraction-backed fallback
commands:
tar -xzf ithz-mcp-ubuntu.tar.gz
cd ithz-mcp-ubuntu
./install.sh
./install_linux_native.sh
ithz-mcp version
ithz-native-resolve
Native Linux project.ithz support uses the bundled build kit and dynamically
loads the system zlib runtime (libz.so.1 / libz.so).
Status
ITHZ MCP is currently free during alpha preview. Public package metadata is published on PyPI and in the official MCP Registry.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.