Purpose
Samuraizer is a local-first CLI tool and MCP server for converting meeting recordings into transcripts, summaries, action items, and decisions. It is the reference implementation of the memnex specification—memnex is an open standard for portable meeting outputs (transcripts, summaries, action items, decisions), and all outputs conform to memnex v0.2 and include a complete provenance chain. All processing is done entirely on the local machine, and recordings never leave your machine.
Key Capabilities
- Fully local operation: Recording data never leaves the local machine, with no cloud services, no subscriptions, and no per-minute billing.
- CLI-first: Scriptable and automatable, and can be integrated with cron, Git hooks, and Obsidian workflows.
- Resumable: If the process crashes midway, rerunning it continues from where it left off.
- Fast: The LLM stages (summaries, action items, decisions) run in parallel, reducing total time by about 2.3x.
- Model-agnostic: Works with any Ollama-compatible LLM, allowing you to choose a suitable model based on your hardware.
- Free: No subscriptions and no per-minute pricing.
- Portable output: Uses the memnex v0.2 format, so other meeting apps, local AI agents, or MCP servers can consume the same transcripts, summaries, action items, decisions, and provenance information without learning Samuraizer's proprietary JSON structure.
- IDE support: memnex has been registered in the Schemastore universal JSON Schema catalog (PR #5676), and *.memnex.json and meeting-output.json files receive autocompletion and validation by default in VS Code, JetBrains IDEs, Neovim, Sublime, and other editors that support language servers.
How to Use
Samuraizer is published as two npm packages, each with its own README and changelog. Specific installation commands are not yet provided.
To verify a memnex document in under a minute, follow the memnex quickstart, which installs the validator and checks a real v0.2 example end to end.
Suitable Scenarios
Suitable for teams or individuals who need to process meeting recordings locally, care about data privacy, and want to automate transcription and summarization workflows. It can be integrated withexisting toolchains (such as cron, Git hooks, and Obsidian).
Limitations and Status
- Specific installation commands are not yet provided.
- The roadmap includes local speaker diarization, encrypted signing UX, accessibility (WebVTT subtitles), and a path to memnex v1.0, all of which are funded milestones.
- Samuraizer follows the BDFL bootstrap governance model, detailed in memnex GOVERNANCE.md, and Samuraizer mirrors that model.
- Bug reports and feature discussions are welcome in GitHub Issues.
- License: MIT, see LICENSE for details.