@clauderecallhq/cli 0.12.1 → 0.61.2

package/README.md

@@ -1,9 +1,9 @@
 <div align="center">
 
 <picture>
-  <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/clauderecallhq/.github/main/profile/assets/wordmark-dark.svg">
-  <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/clauderecallhq/.github/main/profile/assets/wordmark-light.svg">
-  <img src="https://raw.githubusercontent.com/clauderecallhq/.github/main/profile/assets/wordmark-dark.svg" alt="Claude Recall" width="400">
+  <source media="(prefers-color-scheme: dark)" srcset="https://clauderecall.com/wordmark-dark.svg">
+  <source media="(prefers-color-scheme: light)" srcset="https://clauderecall.com/wordmark-light.svg">
+  <img src="https://clauderecall.com/wordmark-dark.svg" alt="Claude Recall" width="400">
 </picture>
 
 <br />
@@ -15,8 +15,13 @@ Claude Recall indexes every Claude Code session you've ever run, surfaces what m
 
 <br />
 
+<img src="https://clauderecall.com/hero/brand-morph.gif" alt="Claude Recall — memory tree morphing into the wordmark" width="700">
+
+<br />
+<br />
+
 [![npm](https://img.shields.io/npm/v/@clauderecallhq/cli?style=flat-square&color=f97316&label=npm)](https://www.npmjs.com/package/@clauderecallhq/cli)
-[![Node](https://img.shields.io/badge/node-%E2%89%A522-43853d?style=flat-square&logo=node.js&logoColor=white)](https://nodejs.org)
+[![Node](https://img.shields.io/badge/node-%E2%89%A520-43853d?style=flat-square&logo=node.js&logoColor=white)](https://nodejs.org)
 [![TypeScript](https://img.shields.io/badge/typescript-strict-3178c6?style=flat-square&logo=typescript&logoColor=white)](https://www.typescriptlang.org/)
 [![Local-first](https://img.shields.io/badge/local--first-127.0.0.1%20only-0b0c0f?style=flat-square)](#privacy--security)
 [![License](https://img.shields.io/badge/license-proprietary-1a1a2e?style=flat-square)](#license)
@@ -101,23 +106,50 @@ Claude picks up where you left off instead of asking you to explain it again.
 <tr>
 <td width="50%" valign="top">
 
-### Search Everything
-Full-text search across 100,000+ messages in milliseconds. BM25 ranking, highlighted snippets, inline `#tag` filters.
+### Context Re-injection &nbsp;<sub>← the moat</sub>
+Pipe any past session straight into a new Claude Code conversation as condensed markdown. Claude picks up where you left off instead of asking you to re-explain it. **Nothing else in this category has it.**
+
+```bash
+recall context <id> | claude
+recall context <id> --prelude "continue this" | claude
+recall context <id> --since 2h | claude
+```
+
+</td>
+<td width="50%" valign="top">
+
+### Threads &nbsp;<sub>NEW · v0.15a</sub>
+Group related sessions into a DAG you can see, rename, and walk. Drag-to-reparent in the graph view. Right-click rename. Bulk-generate coherent titles across an entire thread with one click.
+
+```bash
+recall thread list
+recall thread create "auth-refactor"
+recall thread add <thread> <session>
+```
+
+</td>
+</tr>
+<tr>
+<td width="50%" valign="top">
+
+### Full-Text Search
+FTS5 BM25 ranking across every message, every session, in milliseconds. Highlighted snippets, inline `#tag` filters, project + date scoping.
 
 ```bash
 recall search "zod schema"
-recall search "auth #auth-fix"
+recall search "auth #auth-fix" -p Pest-Control
 ```
 
 </td>
 <td width="50%" valign="top">
 
-### Context Re-injection
-The core feature. Pipe any past session into a new Claude Code conversation as condensed markdown. Runs locally, never touches the network.
+### On-Device Vector Search &nbsp;<sub>Pro</sub>
+Local **768-dimension embeddings** via `bge-base-en-v1.5` ONNX. Three-lane RRF fusion (BM25 + summary + vectors) finds sessions by *meaning*, not keywords. Your code never leaves your laptop.
+
+The embedder model (~110MB) auto-downloads on `npm install`. Skip with `RECALL_SKIP_MODEL_DOWNLOAD=1`.
 
 ```bash
-recall context <id> | claude
-recall context <id> --prelude "continue this" | claude
+recall similar <session-id>
 ```
 
 </td>
@@ -130,8 +162,7 @@ Per-session and per-project token + dollar totals. Daily sparkline. Top-10 heavi
 
 ```bash
 recall stats <id>
-recall stats --project Tools
-recall stats --days 7
+recall stats --project Tools --days 7
 ```
 
 </td>
@@ -150,19 +181,14 @@ recall blame <sha>
 <tr>
 <td width="50%" valign="top">
 
-### Semantic Search
-Conceptual search powered by your local `claude` CLI. Find "the session where I debugged the worker memory leak" without remembering exact words.
-
-```bash
-recall semantic on
-recall semantic backfill
-```
+### Auto-Tagging
+Let Claude label your sessions automatically. Two modes: **MCP mode** (zero-setup, runs through your existing Claude Code subscription at no extra cost) or **BYOK** (paste your own Anthropic API key). Off by default, fully opt-in.
 
 </td>
 <td width="50%" valign="top">
 
-### MCP Server
-Expose Recall to Claude Desktop, Claude Code, or any MCP client as native tools. Read-only by default, opt-in write tools with rate limiting and audit logging.
+### MCP Server &nbsp;<sub>31 tools</sub>
+Expose Recall to Claude Desktop, Claude Code, or any MCP client as native tools. **13 read tools always on**, **18 write tools** opt-in with rate limiting and append-only audit logging.
 
 ```bash
 recall mcp
@@ -174,14 +200,36 @@ recall mcp --allow-writes
 <tr>
 <td width="50%" valign="top">
 
-### Collections & Tags
-Hand-curated hierarchical groupings of sessions that cut across projects. Human-readable aliases. Auto-tagging via your Anthropic API key or Claude CLI (zero extra cost).
+### Collections, Tags & Notes
+Hand-curated hierarchical collections that cut across projects. Flat tags. Markdown notes per session. Aliases. Pins. Cross-session diff. Every write is reversible -- three-layer durability with plain-text mirrors on disk.
 
 </td>
 <td width="50%" valign="top">
 
 ### Web UI
-Beautiful dark-mode three-pane layout. Virtualized transcript rendering (10,000-message sessions open instantly). `Cmd+K` command palette. Multi-select. Export to markdown, HTML, or PDF.
+Dark-mode three-pane layout. **Virtualized transcript** (10,000-message sessions open instantly). `Cmd+K` command palette. Cross-session diff. First-60-seconds onboarding tour. Export to markdown, HTML, or print-ready PDF.
+
+</td>
+</tr>
+<tr>
+<td width="50%" valign="top">
+
+### Rediscovery &nbsp;<sub>"For you"</sub>
+Three rotating cards in the left pane surface forgotten sessions worth revisiting: a 60+-day-old session that overlaps with what you've been working on, your top-cost session this week, and the session that authored your current `HEAD`.
+
+```bash
+recall digest
+```
+
+</td>
+<td width="50%" valign="top">
+
+### IDE Extension
+**VS Code · Cursor · Windsurf · VS Code Insiders.** Tag terminal tabs from inside your editor -- terminal names sync to Recall so sessions get meaningful aliases with zero CLI ceremony, live as you rename tabs.
+
+```bash
+recall install-extension
+```
 
 </td>
 </tr>
@@ -190,16 +238,17 @@ Beautiful dark-mode three-pane layout. Virtualized transcript rendering (10,000-
 <br />
 
 <details>
-<summary><strong>More: VS Code extension, cross-session diff, onboarding, keyboard shortcuts...</strong></summary>
+<summary><strong>More: secrets redaction, memory health, audit log, keyboard shortcuts...</strong></summary>
 
 <br />
 
-- **VS Code / Cursor / Windsurf extension** -- auto-names sessions using terminal tab names via process-tree matching
-- **Cross-session diff** -- split view of two transcripts with shareable `cmp=` URL
-- **First-run onboarding** -- a 3-step tour that demos the moat in the first 60 seconds
-- **Multi-select sessions** -- `Cmd+Click` to toggle, `Shift+Click` for range, right-click to copy IDs
-- **Export** -- condensed markdown, full markdown, self-contained dark-mode HTML, print-ready PDF
-- **Cmd+K command palette** -- jump to any session, action, or setting in one keystroke
+- **Secrets redaction** -- API keys, auth tokens, passwords, private-key blobs are detected by pattern and replaced with placeholders **before** text lands in the searchable index or any export. Original JSONLs untouched.
+- **Memory health scores** -- every session gets a freshness score; surfaces stale data so you know when to re-run a backfill.
+- **MCP audit log** -- every write tool call lands in `mcp_audit_events` with status (ok / error / rate_limited). Pure local, no telemetry.
+- **First-60-seconds onboarding** -- 3-step modal demos the moat (`recall context <id> | claude`) the moment you open the web UI.
+- **Cmd+K command palette** -- jump to any session, action, or setting in one keystroke.
+- **Multi-select sessions** -- `Cmd+Click` toggle, `Shift+Click` range, right-click to copy IDs.
+- **Three-layer durability** -- every write goes to SQLite + an append-only history column + a plain-text mirror on disk. Source JSONLs at `~/.claude/projects/` are strictly read-only.
 
 #### Keyboard Shortcuts
 
@@ -242,6 +291,19 @@ recall open
 
 The daemon watches `~/.claude/projects/` for new and changed session files, indexes them into a local SQLite database at `~/.recall/db.sqlite`, and serves the web UI on a random `127.0.0.1` port.
 
+### Browse
+
+Two interactive surfaces, both work without the daemon (they read SQLite directly):
+
+```bash
+recall                    # smart welcome dashboard with status + slash command palette
+recall tui                # full terminal UI: arrow-key browse, live search, preview pane
+```
+
+`recall` (no args) prints a branded splash with version / daemon / sessions / license status and drops into a slash-aware prompt; type `/` to browse 60+ commands grouped by purpose (Setup, Browse, Pipe to Claude, Threads, Inference, Analytics, Daemon, Diagnostics, Integrations, Sharing, Feedback, Pro).
+
+`recall tui` mounts an Ink-based terminal app modeled on `lazygit` and `k9s`. Arrow keys to navigate, `/` to filter, `enter` to view the full transcript, `o` to open the selected session in the browser, `q` to quit.
+
 ### Explore
 
 ```bash
@@ -265,6 +327,44 @@ recall context abc12345 --prelude "continue where we left off" | claude
 recall context abc12345 --since 2h | claude
 ```
 
+### How Claude Recall integrates with Claude Code
+
+Don't think of Claude Recall as a *separate* tool you context-switch to. It's **Claude Code's memory layer** — you reach it from inside any session through four surfaces, in increasing power:
+
+**1. Bash tool — always available, zero setup.**
+
+Inside any Claude Code conversation, just say *"run `recall search auth bug`"* (or any other recall command). The Bash tool executes it; output lands in the conversation. Works the moment you `npm install -g @clauderecallhq/cli` — no config.
+
+**2. Slash commands (skills) — one-tap workflows.**
+
+When Claude Recall is installed as a Claude Code plugin, typing `/` in Claude Code surfaces ready-made workflows:
+
+| Slash command | What it does |
+|---|---|
+| `/claude-recall:recall <query>` | Search past sessions and pull relevant ones in as context |
+| `/claude-recall:context <id>` | Load one specific session's full transcript |
+| `/claude-recall:neighborhood <id>` | Bundle parents + children + citations + similar |
+| `/claude-recall:tag <id> <tag>` | Tag a session for later retrieval |
+| `/claude-recall:alias <id> <name>` | Rename a session |
+| `/claude-recall:threads-sync` | Capture currently active terminals into a thread |
+| `/claude-recall:thread-scan` | Auto-detect threads across historical sessions |
+| `/claude-recall:setup` | Verify installation and MCP connection |
+
+**3. MCP server — the model calls Recall directly.**
+
+The most powerful surface. With `recall mcp` registered as an MCP server in Claude Code, the model itself decides when to call Recall mid-conversation. Ask *"have I solved this auth bug before?"* and Claude calls `search` autonomously, gets results, and uses them. Ask *"pull yesterday's context"* and it calls `context_for_session`. No commands, no slash menu — pure intent.
+
+**4. Piping into a fresh `claude` — start tomorrow with yesterday's context.**
+
+The cross-session continuation pattern, run from your shell:
+
+```bash
+recall neighborhood abc12345 | claude   # bundle related sessions, start fresh chat
+recall context abc12345 | claude        # one specific session, start fresh chat
+```
+
+Use case: you finish work Monday night, restart Tuesday morning with a clean Claude Code window, and want it to know what happened yesterday without re-explaining.
+
 ### MCP Server
 
 Point any MCP client at Recall and the model calls it as native tools:
@@ -280,11 +380,11 @@ Point any MCP client at Recall and the model calls it as native tools:
 ```
 
 <details>
-<summary><strong>Available MCP tools (31 total)</strong></summary>
+<summary><strong>Available MCP tools (33 total)</strong></summary>
 
 <br />
 
-**Read tools (13 -- always available):**
+**Read tools (14 -- always available):**
 
 | Tool | Purpose |
 |---|---|
@@ -298,11 +398,12 @@ Point any MCP client at Recall and the model calls it as native tools:
 | `semantic_status` | Health snapshot: model, worker, queue depth (Pro) |
 | `get_session` | Full transcript (accepts 8+ char id prefix) |
 | `context_for_session` | Condensed markdown export, ready to inject |
+| `recall_neighborhood` | Bundle parents/children/citations/similar as one budget-bounded markdown blob |
 | `thread_list` | All threads, newest first; can exclude archived |
 | `thread_get` | Full thread detail with every session edge |
 | `thread_for_session` | Non-archived threads referencing a session |
 
-**Write tools (18 -- opt-in via `--allow-writes`):**
+**Write tools (19 -- opt-in via `--allow-writes`):**
 
 | Tool | Purpose |
 |---|---|
@@ -321,6 +422,7 @@ Point any MCP client at Recall and the model calls it as native tools:
 | `thread_archive` | Soft-delete thread; hidden by default |
 | `thread_merge` | Move all edges from source to dest, delete source |
 | `thread_split` | Peel sessions into new thread |
+| `sync_active_sessions` | Capture currently-active terminals in a repo into a thread (Captain Code Pattern) |
 | `generate_thread_titles` | Generate coherent titles for sessions in thread DAG |
 
 All writes are rate-limited (default 60/min), zod-validated, and audited to `~/.recall/audit/`.
@@ -336,27 +438,34 @@ All writes are rate-limited (default 60/min), zod-validated, and audited to `~/.
 ## How It Works
 
 ```
-~/.claude/projects/**/*.jsonl   <-- Claude Code's session files (read-only)
+~/.claude/projects/**/*.jsonl   <-- Claude Code's session files (read-only to Recall)
         |
         v  chokidar file watcher
-~/.recall/db.sqlite             <-- indexed database (FTS5 full-text search)
+~/.recall/db.sqlite             <-- indexed database (FTS5 + sqlite-vec 768d)
+        |
+        +-- background embedding worker (idle-aware, transactional)
+        +-- git correlator (read-only `git log` scoped to session cwd)
+        +-- semantic / auto-tag scanner (opt-in, off by default)
         |
   +-----+-----+
   |           |
   CLI         HTTP server (Hono, 127.0.0.1 only)
   |           |
-  recall ...  Web UI (React + Tailwind SPA)
+  recall ...  Web UI (React + Tailwind SPA, Vite-bundled)
+              MCP server (stdio, 31 tools, opt-in writes)
 ```
 
 ### Data stays local
 
 | Path | What |
 |---|---|
-| `~/.recall/db.sqlite` | Indexed database (projects, sessions, messages, FTS5, aliases, notes, tags, collections, semantic summaries, commit links, usage rollups, audit logs) |
+| `~/.recall/db.sqlite` | Indexed database (projects, sessions, messages, FTS5, vectors, aliases, notes, tags, collections, threads, semantic summaries, commit links, usage rollups, audit logs) |
 | `~/.recall/aliases.json` | Plain-text mirror of every alias with full edit history |
 | `~/.recall/notes/<session>.md` | One markdown file per session note |
 | `~/.recall/tags.json` | Current tag state + append-only event log |
 | `~/.recall/collections.json` | Current tree state + append-only event log |
+| `~/.recall/semantic/<id>.json` | Per-session summary + keywords (semantic mode) |
+| `~/.recall/license.jwt` | Activated Pro license (offline-validatable, dead-man clause) |
 
 <br />
 
@@ -368,8 +477,16 @@ All writes are rate-limited (default 60/min), zod-validated, and audited to `~/.
 
 Claude Recall is **local-first by design**, not as an afterthought.
 
-- The daemon binds to **`127.0.0.1` only**. Never `0.0.0.0`, never a public interface.
-- **No telemetry.** No analytics. No outbound network calls. The only exception is a one-time license check when you activate a Pro key.
+- **The daemon binds to `127.0.0.1` only.** Never `0.0.0.0`, never a public interface.
+- **No telemetry. No analytics. No background phone-home.** The daemon, file watcher, indexer, and web UI never make outbound network calls. Your session content never leaves your machine.
+- **The only outbound calls are user-initiated**, never automatic, and never transmit your session data:
+
+  | Command | Endpoint | What it sends | What it gets |
+  |---|---|---|---|
+  | `recall activate <key>` | `clauderecall.com` | The license key only | A signed JWT for offline validation |
+  | `recall semantic install` | `huggingface.co` | Nothing about you | The `bge-base-en-v1.5` ONNX model file |
+
+  Both are optional and not required for core indexing, search, or context re-injection. Skip them and Recall is fully air-gapped.
 - **No cloud. No account.** Your data lives at `~/.recall/` and nowhere else.
 - **Source sessions are never modified.** Claude Code's JSONL files at `~/.claude/projects/` are strictly read-only to Recall.
 - **Never-delete-data invariant.** Every write enforces three independent durability layers: SQLite with history columns, plain-text mirror on disk, and source JSONLs untouched. Renaming, clearing, or removing anything is always additive -- the prior value is archived, never destroyed.
@@ -388,6 +505,10 @@ Claude Recall is **local-first by design**, not as an afterthought.
 <br />
 
 ```bash
+# Surfaces
+recall                    # smart welcome dashboard + slash command palette (no args)
+recall tui                # interactive terminal UI for browsing and search
+
 # Lifecycle
 recall start              # start daemon (watcher + local HTTP)
 recall stop               # stop daemon
@@ -433,14 +554,38 @@ recall stats --backfill        # one-shot usage backfill
 recall correlate               # link sessions to commits
 recall blame <sha>             # commit -> session reverse lookup
 
-# Semantic search
-recall semantic on             # enable conceptual search
-recall semantic status         # backfill progress
-recall semantic backfill       # run now
+# Semantic / vector search (Pro)
+# Embedder model auto-installs on `npm install`; manual install also available.
+recall semantic install        # download on-device embedding model (auto-runs on npm install)
+recall semantic status         # model + backfill progress + auto-extract state
+recall semantic reindex        # re-embed everything
+recall semantic auto-extract on  # daemon nibbles un-extracted sessions through Claude (Pro)
+recall similar <id>            # cosine kNN over session chunks
+
+# Threads (v0.15a)
+recall thread list             # all threads, newest first
+recall thread create <name>    # new thread
+recall thread add <t> <s>      # attach session to thread
+
+# Rediscovery
+recall digest                  # today's "For you" picks
+recall digest --json           # machine-readable
 
 # MCP server
 recall mcp                     # stdio, read-only
 recall mcp --allow-writes      # opt-in write tools
+
+# Pro license
+recall activate <key>          # one-time activation, offline forever
+recall license                 # show license status
+
+# IDE extension
+recall install-extension       # auto-install for VS Code / Cursor / Windsurf
+
+# Maintenance
+recall health                  # memory health scores per project
+recall paste                   # archive clipboard content into Recall (opt-in)
+recall audit-secrets           # scan index for residual secrets
 ```
 
 </details>
@@ -455,13 +600,16 @@ recall mcp --allow-writes      # opt-in write tools
 
 | Layer | Technology |
 |---|---|
-| CLI + daemon | TypeScript 5 (strict), Node 22+, ESM |
+| CLI + daemon | TypeScript (strict), Node 20+, ESM |
 | HTTP server | Hono + @hono/node-server |
 | Database | better-sqlite3 with FTS5 full-text search |
+| Vector store | sqlite-vec (768d, on-device) |
+| Embeddings | @huggingface/transformers (`bge-base-en-v1.5` ONNX) |
 | File watcher | chokidar |
 | CLI parsing | commander |
 | MCP server | @modelcontextprotocol/sdk (stdio) |
 | Validation | zod |
+| Auth | jose (offline-validatable signed JWT licenses) |
 | Web UI | Vite 8 + React 19 + Tailwind CSS |
 | Data fetching | @tanstack/react-query |
 | Virtualization | @tanstack/react-virtual |
@@ -473,80 +621,136 @@ recall mcp --allow-writes      # opt-in write tools
 
 <br />
 
+## Supported Platforms
+
+Claude Recall is a Node.js CLI with a few native dependencies (`better-sqlite3`, `sqlite-vec`, and the optional `@huggingface/transformers` for semantic search). npm picks the right prebuilt binary for your platform automatically; no compile toolchain required.
+
+| Platform | Tier | Status |
+|---|---|---|
+| **macOS** arm64 (Apple Silicon, M1+) | Tier 1 | Tested in pre-publish smoke test |
+| **macOS** x64 (Intel) | Tier 1 | Tested in pre-publish smoke test |
+| **Linux** x64 (glibc, kernel 4.x+) | Tier 1 | Tested on every CI run (Node 22, 24) |
+| **Linux** arm64 (glibc) | Tier 2 | Best-effort. Native deps ship arm64 prebuilds; report issues if you hit one |
+| **Windows** x64 (Windows 10+) | Tier 1 | Tested in pre-publish smoke test |
+| Windows arm64 / Linux musl (Alpine) / 32-bit | Unsupported | Native deps lack reliable prebuilds; install will be rejected |
+
+**Node:** 22 LTS or 24 LTS. Node 20 and earlier are unsupported (declared in `engines.node`).
+
+**Semantic search is optional.** The on-device embedder is the only feature that loads the native ONNX runtime. The model auto-downloads (~110MB) on `npm install`; skip with `RECALL_SKIP_MODEL_DOWNLOAD=1` and install later via `recall semantic install`. Core CLI features — search, list, context, daemon, MCP — work on every supported platform regardless of whether the embedder is installed. If the embedder fails to load on your platform, you get a clear error pointing here, and the rest of Claude Recall keeps working.
+
+<br />
+
+---
+
+<br />
+
 ## Distribution
 
 Claude Recall ships on every surface a developer already works in.
 
-| Surface | Identity | Version | Install |
-|---|---|---|---|
-| **npm** (CLI + MCP) | [`@clauderecallhq/cli`](https://www.npmjs.com/package/@clauderecallhq/cli) | v0.12.0 | `npm i -g @clauderecallhq/cli` |
-| **VS Code Marketplace** | [`clauderecallhq.clauderecall-vscode`](https://marketplace.visualstudio.com/items?itemName=clauderecallhq.clauderecall-vscode) | v0.8.3 | Search "Claude Recall" in Extensions |
-| **MCP Registry** | [`io.github.clauderecallhq/recall`](https://registry.modelcontextprotocol.io/v0.1/servers/io.github.clauderecallhq%2Frecall/versions/0.12.0) | v0.12.0 | Auto-discovered by MCP clients |
-| **Plugin Hub** | `claude-recall` | -- | `/plugin install claude-recall` in Claude Code |
-| **Website** | [clauderecall.com](https://clauderecall.com) | v0.4.0 | -- |
+| Surface | Identity | Install |
+|---|---|---|
+| **npm** (CLI + MCP) | [`@clauderecallhq/cli`](https://www.npmjs.com/package/@clauderecallhq/cli) | `npm i -g @clauderecallhq/cli` |
+| **VS Code Marketplace** | [`clauderecallhq.clauderecall-vscode`](https://marketplace.visualstudio.com/items?itemName=clauderecallhq.clauderecall-vscode) | Search "Claude Recall" in Extensions |
+| **MCP Registry** | [`io.github.clauderecallhq/recall`](https://registry.modelcontextprotocol.io/) | Auto-discovered by MCP clients |
+| **Claude Code Plugin Hub** | `claude-recall` | `/plugin install claude-recall` in Claude Code |
+| **Website** | [clauderecall.com](https://clauderecall.com) | -- |
 
-### How the pieces connect
+The npm package ships **the CLI and the MCP server**. The VS Code extension is a thin companion that talks to the local daemon over `127.0.0.1`. The MCP Registry and Claude Code Plugin Hub both point back to the npm package -- one source of truth, one update pipeline.
 
-```
-npm package (@clauderecallhq/cli)
-  |
-  +-- CLI binary: `recall`
-  +-- MCP server: `recall mcp` (stdio transport)
-  |
-  +-- MCP Registry points here via npx
-  +-- Plugin Hub points here via npx
-  |
-VS Code extension (independent)
-  +-- Talks to daemon over HTTP (127.0.0.1)
-  +-- Does NOT run MCP tools
-  |
-Website (clauderecall.com)
-  +-- Auto-deploys on push via GitHub Actions
-  +-- Marketing, docs, /metrics dashboard
-```
+<br />
 
-### Tracking installs
+---
+
+<br />
+
+## Pricing
+
+One tier is free forever. One tier is a **one-time** purchase that works offline for life, even if we disappear. One tier is a conversation.
+
+<table>
+<tr>
+<td width="33%" valign="top">
+
+### Free
+
+**$0**
+Forever. No signup. No credit card.
+
+- Full indexing of every `~/.claude/projects/` JSON-L
+- CLI: `list`, `show`, `projects`, `status`
+- Web UI: browse projects, read transcripts, copy code
+- Three-layer durability
 
 ```bash
-recall installs           # npm downloads + VS Code installs in one view
-recall installs --json    # machine-readable output
+npm install -g @clauderecallhq/cli
 ```
 
-Live dashboard: [clauderecall.com/metrics](https://clauderecall.com/metrics)
+</td>
+<td width="33%" valign="top">
 
-### Publishing checklist (for maintainers)
+### Pro &nbsp;<sub>← daily driver</sub>
 
-1. **npm**: `npm publish --access public` (2FA required, bin entries show warning but work via `npx -p @clauderecallhq/cli recall`)
-2. **VS Code**: Build .vsix in `extensions/vscode/`, upload at [marketplace.visualstudio.com/manage](https://marketplace.visualstudio.com/manage/publishers/clauderecallhq)
-3. **MCP Registry**: `cd plugin && mcp-publisher publish` (requires GitHub OAuth, org membership must be Public)
-4. **Plugin Hub**: Web form at `claude.ai/settings/plugins/submit`
-5. **Website**: Push to `main` -- auto-deploys
+**$29.69 one-time** through May 2026 (Founder pricing)
+**$49.69 one-time** from June 2026 (standard)
+Pay once. Offline forever. Lifetime updates. Dead-man clause included. Never a subscription.
 
-### Known issue: npx scoped bin resolution
+- Full-text search (FTS5 BM25)
+- On-device semantic vector search (768d, RRF fusion)
+- `recall context` re-injection, **the moat**
+- MCP server (Claude Desktop / Claude Code native tools)
+- VS Code / Cursor / Windsurf extension
+- Aliases, tags, markdown notes, pins, date filters, cross-session diff
+- Export: clean markdown, self-contained dark-mode HTML, print-ready
+- Virtualized transcript (10k+ message sessions open instantly)
+- Auto-tagging (API-key + MCP modes)
+- Secrets redaction (client-side scrub of keys, tokens, passwords pre-index)
+- **Every future personal-use feature.** One price, lifetime updates.
 
-`npx -y @clauderecallhq/cli mcp` fails because npx looks for a bin named `cli` (the unscoped package name). Workaround: `npx -p @clauderecallhq/cli recall mcp`. The MCP config in docs uses the workaround syntax.
+[**Buy Pro for $29.69 (Founder pricing through May) →**](https://clauderecall.com/pricing)
+
+</td>
+<td width="33%" valign="top">
+
+### Enterprise
+
+**Contact us**
+Built for 50+ developer organizations.
+
+- Shared session library with end-to-end encrypted team sync
+- Team-wide tag conventions
+- Local audit log (compliance-grade, not surveillance)
+- SSO + per-project access control
+- Priority support, named engineer
+- Optional on-prem deployment
+
+[hello@clauderecall.com](mailto:hello@clauderecall.com)
+
+</td>
+</tr>
+</table>
 
 <br />
 
----
+### How it compares to what you already pay
+
+| Tool | Price |
+|---|---|
+| **Claude Recall** | **$29.69 one-time** through May 2026, **$49.69 one-time** from June |
+| Claude Code | $20 / mo |
+| Cursor | $20 / mo |
+| Raycast Pro | $8 to $16 / mo |
+| GitHub Copilot | $10 / mo |
+
+<sub>Prices as of May 2026.</sub>
 
 <br />
 
-## Pricing
+### The dead-man clause
 
-| | Free | Pro |
-|---|---|---|
-| **Search** | Unlimited | Unlimited |
-| **Context re-injection** | Unlimited | Unlimited |
-| **Sessions indexed** | Unlimited | Unlimited |
-| **Cost analytics** | Unlimited | Unlimited |
-| **Git correlation** | Unlimited | Unlimited |
-| **MCP server** | Read-only | Read + Write |
-| **Semantic search** | -- | Included |
-| **Auto-tagging** | -- | Included |
-| **Price** | $0 | **$29.69 one-time** |
+Your Pro license validates against our server exactly once, at activation. After that, Recall works **offline forever**. If `clauderecall.com` ever disappears, every existing license stays valid -- we'll publish the validation secret if we ever shut down, so you're never stranded.
 
-No subscriptions. No recurring fees. Pay once, own it forever.
+Full pricing FAQ: **[clauderecall.com/pricing](https://clauderecall.com/pricing)**
 
 <br />
 
@@ -563,15 +767,21 @@ No subscriptions. No recurring fees. Pay once, own it forever.
 - [x] VS Code / Cursor / Windsurf extension
 - [x] MCP server (read-only)
 - [x] Collections (hierarchical, cross-project)
-- [x] Auto-tagging
+- [x] Auto-tagging (MCP + BYOK modes)
 - [x] Cost / token analytics
 - [x] Git correlation
-- [x] Semantic search
 - [x] Cmd+K command palette
-- [x] Bidirectional MCP write tools
-- [x] First-run onboarding
+- [x] Bidirectional MCP write tools (rate-limited + audited)
+- [x] First-60-seconds onboarding
+- [x] Rediscovery surface ("For you" cards)
+- [x] **On-device vector search** (768d ONNX embeddings, RRF fusion)
+- [x] **Threads** -- DAG view, drag-to-reparent, bulk title generation
+- [x] **Pro license + dead-man clause** (offline-validatable JWT)
+- [x] Memory health scores
+- [x] Claude Code plugin manifest
 - [ ] Public launch at clauderecall.com
-- [ ] Team features
+- [ ] Team features (E2E-encrypted sync, SSO, audit log)
+- [ ] Enterprise tier (on-prem, named-engineer support)
 
 <br />
 
@@ -582,8 +792,8 @@ No subscriptions. No recurring fees. Pay once, own it forever.
 <div align="center">
 
 <picture>
-  <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/clauderecallhq/.github/main/profile/assets/mark-scrollback-dark.svg">
-  <img src="https://raw.githubusercontent.com/clauderecallhq/.github/main/profile/assets/mark-scrollback-dark.svg" alt="Claude Recall" width="48">
+  <source media="(prefers-color-scheme: dark)" srcset="https://clauderecall.com/mark-scrollback-dark.svg">
+  <img src="https://clauderecall.com/mark-scrollback-dark.svg" alt="Claude Recall" width="48">
 </picture>
 
 <br />
@@ -591,10 +801,10 @@ No subscriptions. No recurring fees. Pay once, own it forever.
 
 **Claude Recall** -- the memory layer for the agent you're steering.
 
-[Website](https://clauderecall.com) &nbsp;&middot;&nbsp; [Install](https://www.npmjs.com/package/@clauderecallhq/cli) &nbsp;&middot;&nbsp; [Issues](https://github.com/clauderecallhq/.github/issues)
+[Website](https://clauderecall.com) &nbsp;&middot;&nbsp; [Install](https://www.npmjs.com/package/@clauderecallhq/cli) &nbsp;&middot;&nbsp; [Issues](https://gitlab.com/clauderecallhq/clauderecallhq/-/issues)
 
 <br />
 
-<sub>Built by <a href="https://github.com/clauderecallhq">@clauderecallhq</a></sub>
+<sub>Built by <a href="https://gitlab.com/clauderecallhq">@clauderecallhq</a></sub>
 
 </div>