$ loading_
Give AI agents deterministic local project memory, context packs, and checkpoints.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "ITHZ MCP" MCP server from askskill: Run: claude mcp add 'dev-ithz-ithz-mcp' -- npx -y ithz-mcp
Use ITHZ MCP to create a local project memory for the current repository, summarize the tech stack, folder structure, run commands, and current development goals, and save them as a reusable context pack.
A structured project memory is created and saved as a context pack for future tasks.
Use ITHZ MCP to create a checkpoint before refactoring, recording the current key file states, known risks, and rollback notes so later changes can be compared and the workspace can be restored.
A traceable checkpoint is created with a pre-change summary and rollback references.
Use ITHZ MCP to set gates for a debugging task: load only the authentication module, recent error logs, and relevant configuration, while excluding unrelated documents and past experiment content.
A filtered minimal task context is produced, reducing noise and improving AI agent reliability.
No documentation provided
Check the source repo for usage and examples.
Give AI coding agents local-first project memory, decisions, risks, and checkpoints.
Give MCP agents local-first long-term memory without databases or API keys.
Provide local-first memory and retrieval for coding agents using SQLite embeddings.
Store project rules, context, and decisions for reuse across MCP clients.
Provide centralized persistent memory, validation, and MCP integration for LLM workflows.
Manage local-first agent memory with versioning, search, ACLs, and sync.