@asobacloud/nehanda 0.2.0 → 0.3.0

package/README.md

@@ -18,22 +18,35 @@ Every conversation turn, tool call, phase transition, and permission check is st
 
 ### 1. Installation
 
+**Global install (recommended):**
+
+```bash
+npm install -g @asobacloud/nehanda
+```
+
+During install, `npm` automatically provisions a dedicated Python venv and downloads the [Laya](https://huggingface.co/convaiinnovations/laya) checkpoint (~808 MB). This happens once — subsequent starts are instant.
+
+**From source:**
+
 ```bash
 git clone https://github.com/AsobaCloud/nehanda-cli.git
 cd nehanda-cli
 npm install
 ```
 
-On first launch, the CLI automatically provisions a dedicated Python venv and downloads the [Laya](https://huggingface.co/convaiinnovations/laya) checkpoint (~808 MB). This happens once. Subsequent starts are instant.
+The same one-time provisioning runs automatically as part of `npm install`.
 
 ### 2. Launching the REPL
 
-Launch the interactive Ink TUI:
+```bash
+nehanda
+```
+
+Or via the aliases installed with the package:
 
 ```bash
-npm start
-# or directly run:
-node bin/nehanda-ui.mjs
+ona
+ona-code
 ```
 
 ### 3. Provider Setup
@@ -322,6 +335,17 @@ To clear a key:
 
 **Tip:** Keys are masked in `/mcp env` output. Only the first 6 and last 4 characters are shown.
 
+#### Connecting to JupyterLab
+
+`mcp.json` ships with a `jupyter` server pre-configured pointing at `localhost:8888`. To connect it to your JupyterLab instance:
+
+```
+❯ /mcp env jupyter JUPYTER_URL http://my-server:8888
+❯ /mcp env jupyter JUPYTER_TOKEN <your-token>
+```
+
+Both values are written to `~/.config/nehanda/mcp.json` and never committed. The server reloads automatically.
+
 ### General Config from the REPL
 
 Use `/config set <dot.path> <value>` to set any configuration value without editing files:
@@ -392,6 +416,22 @@ node tests/unit/orchestrate_jev_hotpath_behavioral.mjs
 
 ---
 
+## Changelog
+
+### 0.3.0
+- Published as `@asobacloud/nehanda` on npm — install with `npm install -g @asobacloud/nehanda`
+- Laya System-1 venv provisioning moved to `npm install` time via `scripts/postinstall.mjs` — CLI starts instantly on every subsequent launch
+- Laya provisioning now streams live pip progress to the terminal instead of running silently
+- Added project-local `mcp.json` with `datalayer/jupyter-mcp-server` preconfigured for JupyterLab — set your server URL and token with `/mcp env jupyter JUPYTER_URL <url>` and `/mcp env jupyter JUPYTER_TOKEN <token>`
+
+### 0.2.0
+- Jev-Mem memory architecture (System-1 Laya control plane)
+- Dynamic Tool Rescue (`[TOOL_CALL]` delimiter path)
+- Deterministic 6-phase SDLC workflow
+- Multi-provider switching
+
+---
+
 ## References
 
 Jiang, Z., Li, Y., & Li, J. (2026). *System-One-Controlled Agentic Memory for Efficient AI Agents*. arXiv:2609.23986. https://arxiv.org/abs/2609.23986

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@asobacloud/nehanda",
-  "version": "0.2.0",
+  "version": "0.3.0",
   "type": "module",
   "description": "Agentic terminal REPL and single-process engine for governed AI deep research and software development. Connects to Nehanda v3, Ollama, LM Studio, or any OpenAI-compatible API.",
   "engines": {