cerebro-code 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Yogesh Maheshwari
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,71 @@
+# 🧠 Cerebro
+
+**See, find, and remember everything your Claude Code agent did.**
+
+Cerebro is a local-first dashboard for your Claude Code session history. It reads
+the sessions already on your machine, maps them, surfaces the ones you lost or left
+unfinished, and (on demand) mines them for reusable skills — all without your code
+ever leaving your laptop unless you explicitly ask it to.
+
+> Status: **v0.1** — the browse-and-search core is real and tested. Clustering and
+> skill-mining are on the roadmap below.
+
+```bash
+npx cerebro-code
+```
+
+That's it. It indexes `~/.claude/projects`, starts a local server, and opens your browser.
+
+---
+
+## Why
+
+If you use Claude Code seriously, sessions pile up fast — scattered across repos,
+some left open for days, most impossible to find again later. Cerebro turns that
+flat pile of `.jsonl` files into something you can actually navigate.
+
+## Features (v0.1)
+
+- **Dashboard** — sessions, repos, cost, and unclosed work at a glance
+- **Session browser** — search and filter by repo, type, and recency
+- **Reader** — a clean, readable view of any past session with its tool-call timeline
+- **Resume** — jump back into any session via `claude --resume`
+- **Unclosed detection** — surfaces sessions you started and never finished
+- **Cost view** — cache-aware per-session and per-repo token cost
+- **Per-session summaries** *(on demand)* — uses your own installed `claude`
+
+## Privacy
+
+Cerebro is **local-first by default**. Your session files — which routinely contain
+secrets, tokens, and `.env` contents — are read and indexed **only on your machine**.
+Nothing is sent anywhere until *you* trigger an analysis, and when you do:
+
+- a **deterministic redaction pass** masks API keys, tokens, and credentials first,
+- you see an **egress preview** of exactly what will be sent, and
+- `--no-llm` mode disables all outbound calls entirely.
+
+Raw session content is never persisted by Cerebro.
+
+## Development
+
+```bash
+git clone https://github.com/Yoge5h9/cerebro-code.git
+cd cerebro-code
+npm install
+npm run build        # build web + server
+npm start            # run the CLI locally
+npm test             # unit tests (vitest)
+npm run test:e2e     # browser tests (playwright)
+```
+
+## Roadmap
+
+- **v0.1** — ingest · dashboard · reader · search · cost · unclosed · summaries ✅
+- **v0.2** — Claude-powered clustering + the visual "map of your work"
+- **v0.3** — skill & memory mining → review-gated `SKILL.md` proposals
+- **v1.0** — single-binary install, team session export
+- **later** — provider-agnostic adapters (Cursor, Codex, Cline)
+
+## License
+
+MIT © Yogesh Maheshwari