@wipcomputer/memory-crystal 0.7.30 → 0.7.32

package/.worktrees/memory-crystal-private--cc-mini-release-notes-v0.7.30/skills/memory/SKILL.md

@@ -1,438 +0,0 @@
----
-name: wip-memory-crystal
-description: Search and manage the shared memory crystal. Use when user says "do you remember", "search memory", "remember this", "forget that", "memory status", "what do you know about", or needs to recall past discussions, store facts, or check what's in memory.
-version: 0.7.24
----
-
-# Memory Crystal
-
-All your AIs share one memory. Searchable, private, sovereign. Every conversation you have with any AI gets captured, embedded, and made searchable. Facts, preferences, and decisions persist across sessions and across tools.
-
-## What it does
-
-1. **Captures** every conversation automatically (cron job, every minute)
-2. **Embeds** conversation chunks for semantic search (OpenAI, Ollama, or Google embeddings)
-3. **Searches** across all your AI conversations with ranked, recency-weighted results
-4. **Stores** facts, preferences, and decisions you want to keep permanently
-5. **Works across** Claude Code CLI, Claude Desktop, OpenClaw, and any MCP-compatible client
-
-## What changes for the user
-
-After install, the AI remembers everything. Past conversations are searchable. Important facts persist. The user never has to re-explain context. Every AI tool they use shares the same memory.
-
----
-
-## Intelligent Install & Update
-
-When installing from Claude Code CLI or OpenClaw, Memory Crystal discovers your existing AI sessions automatically. Installs the Learning Dreaming Machines Operating System (LDM OS) and creates a living memory system. From this point forward, every conversation is captured, archived, and embedded into searchable memory. Choose to install as Crystal Core (all your memories) or Crystal Node (a Crystal Core mirror).
-
-If already installed, `crystal init` detects the current version and updates in place. Code gets updated. Data never gets touched.
-
-## Install
-
-**You are an AI agent helping a user install Memory Crystal. Walk them through the process step by step. Ask questions, explain what you're doing, and let them make choices.**
-
-### Question 1: Developer or end user?
-
-Ask the user:
-
-> "Are you installing this as a developer, or do you just want it to work?"
-
-**Developer path:**
-- Fork the repo so they can contribute back
-- Clone locally, build from source
-- They end up with `crystal` and `crystal-mcp` on PATH via `npm link`
-
-```bash
-git clone https://github.com/<their-fork>/memory-crystal.git
-cd memory-crystal
-npm install
-npm run build
-npm link
-```
-
-**End user path:**
-- One command, done
-
-```bash
-npm install -g @wipcomputer/memory-crystal
-```
-
-If npm is not available, check if the user has Node.js installed (`node --version`). They need Node.js 20+. If not installed, tell them to install it from https://nodejs.org first.
-
-Both paths end with `crystal` and `crystal-mcp` on PATH.
-
-### Question 2: First install or adding a device?
-
-Ask the user:
-
-> "Is this your first time installing Memory Crystal, or do you already have it on another machine?"
-
-**First install:** Full setup. This machine becomes the source of truth. Continue to Step 1.
-
-**Adding a device:** They already have a Crystal Core somewhere. Skip to "Adding a Device" below.
-
-### Transparency: Tell the user what's about to happen
-
-Before installing, explain exactly what will be created:
-
-> "Here's what Memory Crystal will set up on your machine:
->
-> - `~/.ldm/` ... a hidden folder in your home directory. This is where everything lives.
-> - `~/.ldm/memory/crystal.db` ... your memory database. All conversations, all memories.
-> - `~/.ldm/agents/` ... per-agent data (transcripts, daily logs, sessions)
-> - `~/.ldm/bin/crystal-capture.sh` ... a script that captures conversations every minute via cron
-> - `~/.ldm/bin/ldm-backup.sh` ... a backup script (optional daily schedule)
->
-> Nothing gets installed outside this folder. Nothing phones home. Want me to go ahead?"
-
-### Step 0: Install or upgrade the global CLI (ALWAYS run this)
-
-**Always run this before crystal init.** Even if crystal is already installed. This ensures the global CLI is the latest version. Never skip this step. Never build from source and copy to extensions instead.
-
-```bash
-npm uninstall -g memory-crystal 2>/dev/null
-npm install -g @wipcomputer/memory-crystal
-```
-
-Then proceed to Step 1.
-
-### Step 1: Initialize
-
-```bash
-crystal init
-```
-
-This does everything: scaffolds `~/.ldm/`, deploys code to `~/.ldm/extensions/memory-crystal/`, configures the Claude Code Stop hook, registers the MCP server, deploys capture and backup scripts, and installs a cron job. If OpenClaw is detected, it deploys the OC plugin too. Safe to run multiple times. If already installed, it detects the version and updates if needed.
-
-You can also specify a role during init:
-
-```bash
-crystal init --core          # Install as Crystal Core
-crystal init --node --pair mc1:...   # Install as Node with pairing code
-```
-
-**Core recommendation:** If this machine is always on (desktop, server, Mac mini), it should be your Crystal Core. The Core is the master memory. It does all embeddings and is the source of truth. If you're on a laptop, you can still install standalone. But when you're ready, you'll want a Core running on something permanent.
-
-### Step 2: Set up embeddings
-
-Memory Crystal needs an embedding provider to make conversations searchable. Check if the user already has one:
-
-```bash
-echo $OPENAI_API_KEY
-```
-
-If that returns a key, skip to Step 3.
-
-If not, ask the user which provider they want:
-
-**Option A: OpenAI (recommended, requires API key)**
-Tell the user to add this to their shell profile (`~/.zshrc` or `~/.bashrc`):
-```bash
-export OPENAI_API_KEY="sk-..."
-```
-They get the key from https://platform.openai.com/api-keys
-
-**Option B: Ollama (free, local, no API key needed)**
-```bash
-ollama --version
-# If not installed: https://ollama.com
-ollama pull nomic-embed-text
-```
-Then add to shell profile:
-```bash
-export CRYSTAL_EMBEDDING_PROVIDER=ollama
-```
-
-**Option C: Google**
-```bash
-export CRYSTAL_EMBEDDING_PROVIDER=google
-export GOOGLE_API_KEY="..."
-```
-
-### Step 3: Connect to your AI
-
-`crystal init` already handled the main connections automatically:
-- **Claude Code CLI:** Stop hook configured in `~/.claude/settings.json`, MCP server registered
-- **OpenClaw:** If detected, plugin deployed to `~/.openclaw/extensions/memory-crystal/`
-
-Verify the connections worked by running `crystal doctor`. If the MCP server or hook checks show warnings, fix them manually:
-
-#### Claude Code CLI (manual fallback)
-
-If `crystal init` couldn't register the MCP server automatically:
-
-```bash
-claude mcp add --scope user memory-crystal -- node ~/.ldm/extensions/memory-crystal/dist/mcp-server.js
-```
-
-Then restart Claude Code (exit and re-open, or run `/mcp` to reconnect).
-
-After restart, you will have these tools: `crystal_search`, `crystal_remember`, `crystal_forget`, `crystal_status`.
-
-#### Claude Desktop (macOS app)
-
-Check if Claude Desktop is installed:
-
-```bash
-ls ~/Library/Application\ Support/Claude/claude_desktop_config.json 2>/dev/null
-```
-
-If it exists, read the file and add the memory-crystal MCP server to it:
-
-```json
-{
-  "mcpServers": {
-    "memory-crystal": {
-      "command": "node",
-      "args": ["~/.ldm/extensions/memory-crystal/dist/mcp-server.js"]
-    }
-  }
-}
-```
-
-Merge this into the existing config (don't overwrite other servers). Tell the user to restart Claude Desktop.
-
-#### OpenClaw (manual fallback)
-
-If `crystal init` didn't deploy to OpenClaw automatically:
-
-```bash
-cp -r ~/.ldm/extensions/memory-crystal/{dist,skills,package.json,openclaw.plugin.json} ~/.openclaw/extensions/memory-crystal/
-cd ~/.openclaw/extensions/memory-crystal && npm install --omit=dev
-openclaw gateway restart
-```
-
-**All runtimes share the same database.** Tell the user: "All your AIs share one memory at `~/.ldm/memory/crystal.db`."
-
-### Step 4: Bridge (AI-to-AI Communication)
-
-Memory Crystal includes Bridge, which lets your AIs talk to each other. Install it:
-
-```bash
-crystal bridge setup
-```
-
-If Bridge isn't installed yet:
-
-```bash
-npm install -g lesa-bridge
-crystal bridge setup
-```
-
-This registers the Bridge MCP server. Your AIs can now send messages to each other and search each other's conversations.
-
-### Step 5: Backups (optional)
-
-Offer to set up automated backups:
-
-> "Want me to set up daily backups of your memory? I'll install a backup that runs at 3 AM and keeps the last 7 copies."
-
-```bash
-crystal backup setup
-```
-
-The backup destination defaults to `~/.ldm/backups/`. The user can change this by setting `LDM_BACKUP_DIR` in their shell profile to wherever they trust: iCloud Drive, an external drive, Dropbox, etc.
-
-**If the Core is on a laptop:** Strongly recommend backups. Laptops get lost, stolen, spilled on. The Core is the source of truth. Back it up.
-
-### Step 6: Verify
-
-Run `crystal doctor` to check that everything is set up correctly:
-
-```bash
-crystal doctor
-```
-
-This shows the status of every component: database, embeddings, capture, relay, MCP, backup, bridge. If anything is wrong, it tells you how to fix it.
-
-Then test search:
-
-```bash
-crystal search "test"
-```
-
-If that works, tell the user: "Memory Crystal is installed. From now on, I can search our past conversations, remember important things, and share memory with your other AI tools. Try asking me 'do you remember what we talked about last week?'"
-
----
-
-## Update
-
-If Memory Crystal is already installed and a new version is available, update it:
-
-> "Update Memory Crystal to the latest version"
-
-The AI runs:
-
-```bash
-crystal update
-```
-
-This detects the installed version, shows what will change, and deploys the new code. It updates:
-- Code in `~/.ldm/extensions/memory-crystal/dist/`
-- Code in `~/.openclaw/extensions/memory-crystal/dist/` (if OpenClaw is present)
-- Skills and package manifests
-- CC Stop hook path (if changed)
-- MCP server registration (if needed)
-
-It never touches:
-- `~/.ldm/memory/crystal.db` (your data)
-- `~/.ldm/state/*` (watermarks, role state)
-- `~/.ldm/secrets/*` (relay key)
-- `~/.ldm/agents/*` (agent data, transcripts, daily logs)
-
-After the update, run `crystal doctor` to verify everything is working. If the update changed hook paths or MCP registration, restart Claude Code.
-
----
-
-## Adding a Device
-
-If the user already has a Crystal Core on another machine:
-
-### Step 1: Install the package
-
-Same as above (developer fork or `npm install -g @wipcomputer/memory-crystal`).
-
-### Step 2: Initialize as a Node
-
-```bash
-crystal init --agent <name>
-```
-
-Use a descriptive agent name like `cc-air`, `cc-laptop`, etc.
-
-### Step 3: Pair with the Core
-
-On the Core machine:
-```bash
-crystal pair
-```
-This shows a QR code and a pairing string.
-
-On this machine:
-```bash
-crystal pair --code mc1:...
-```
-
-Both machines now share the encryption key.
-
-### Step 4: Configure relay
-
-Ask the user: "Do you want to use the free WIP.computer relay, or set up your own?"
-
-**Option A: WIP.computer relay (recommended)**
-- Free during beta. Nothing to set up
-- Your data is end-to-end encrypted. The relay is blind
-- Set env vars:
-```bash
-export CRYSTAL_RELAY_URL=<provided-url>
-export CRYSTAL_RELAY_TOKEN=<provided-token>
-export CRYSTAL_AGENT_ID=<agent-name>
-```
-
-**Option B: Self-hosted relay (full sovereignty)**
-- Deploy your own Cloudflare Worker + R2 bucket
-- Requires a Cloudflare account (free tier works)
-- Walk them through the setup in RELAY.md
-
-### Step 5: Connect to AI + Bridge + Verify
-
-Same as first install Steps 3-6 above.
-
-### Step 6: Demote to Node
-
-```bash
-crystal demote
-```
-
-This machine is now a Crystal Node. Conversations are captured, encrypted, and relayed to the Core. The Core embeds everything and pushes a searchable mirror back.
-
----
-
-## Role Management
-
-Users can check and change roles at any time:
-
-```bash
-crystal role              # Show current role
-crystal promote           # Make this device the Crystal Core
-crystal demote            # Make this device a Crystal Node
-```
-
-If a user starts on a laptop and later gets a desktop, they can promote the desktop and demote the laptop. No data loss.
-
----
-
-## Coming Back Later
-
-Users can always come back and say:
-
-> "Hey, can you check what Memory Crystal features I have installed and what I'm missing?"
-
-Run:
-
-```bash
-crystal doctor
-```
-
-This shows the full state of the install: role, database, embeddings, capture, relay, MCP, backup, bridge. Each check shows OK, WARN, or FAIL with a suggested fix.
-
----
-
-## Tools
-
-Once installed, these tools are available to the AI:
-
-### crystal_search
-
-Search across all stored memory. Semantic search with recency-weighted results.
-
-```
-crystal_search query="how do plugins work" limit=5
-crystal_search query="user preferences" agent_id="main"
-```
-
-Results are ranked by relevance and freshness, with color-coded freshness indicators:
-- fresh (less than 3 days)
-- recent (less than 7 days)
-- aging (less than 14 days)
-- stale (14+ days)
-
-### crystal_remember
-
-Store a fact or observation that persists across sessions.
-
-```
-crystal_remember text="User prefers Opus for complex tasks" category="preference"
-crystal_remember text="API key rotated on 2026-03-01" category="event"
-```
-
-Categories: fact, preference, event, opinion, skill
-
-### crystal_forget
-
-Deprecate a stored memory by ID (marks as deprecated, doesn't delete).
-
-```
-crystal_forget id=42
-```
-
-### crystal_status
-
-Check memory health: chunk count, agents, provider, data directory.
-
-```
-crystal_status
-```
-
----
-
-## Tips
-
-- Search is semantic. "how do plugins work" finds conversations about plugin architecture even if those exact words weren't used.
-- Store preferences and decisions as memories. They survive compaction and context limits.
-- Use `agent_id` filter when you only want results from a specific agent.
-- The cron job captures conversations every minute. No data loss even in long sessions.
-- Available providers: openai (default), ollama (local, free), google.
-- All runtimes (Claude Code, Claude Desktop, OpenClaw) share the same database at `~/.ldm/memory/crystal.db`.
-- Run `crystal doctor` anytime to check what's installed and what's missing.

package/.worktrees/memory-crystal-private--cc-mini-release-notes-v0.7.30/wrangler-demo.toml

@@ -1,8 +0,0 @@
-name = "memory-crystal-demo"
-main = "dist/worker-demo.js"
-compatibility_date = "2024-12-01"
-compatibility_flags = ["nodejs_compat"]
-
-[[kv_namespaces]]
-binding = "DEMO_KV"
-id = "7dab6efe89f443688e178f6135aa3f3b"

package/.worktrees/memory-crystal-private--cc-mini-release-notes-v0.7.30/wrangler-mcp.toml

@@ -1,24 +0,0 @@
-name = "memory-crystal-cloud"
-main = "dist/worker-mcp.js"
-compatibility_date = "2024-12-01"
-
-# D1 — OAuth tables + cloud storage (chunks, memories)
-[[d1_databases]]
-binding = "DB"
-database_name = "memory-crystal-cloud"
-database_id = "40ca6b73-3701-453e-adb3-7faf1a9964ad"  # fill after: wrangler d1 create memory-crystal-cloud
-
-# Vectorize — Semantic vector search (1024 dims, cosine)
-[[vectorize]]
-binding = "VECTORIZE"
-index_name = "memory-crystal-chunks"
-
-# R2 — Shared relay bucket (for Tier 1 sovereign drops)
-[[r2_buckets]]
-binding = "RELAY"
-bucket_name = "memory-crystal-relay"
-
-# Secrets (set via `wrangler secret put --config wrangler-mcp.toml`):
-#   OPENAI_API_KEY          — for embeddings (text-embedding-3-small)
-#   RELAY_ENCRYPTION_KEY    — base64, for Tier 1 relay encryption
-#   MCP_SIGNING_KEY         — for signing OAuth tokens

package/_trash/RELEASE-NOTES-v0-7-23.md

@@ -1,48 +0,0 @@
-# Release Notes: Memory Crystal v0.7.23
-
-**Date:** 2026-03-15
-
-## Search Quality v2 + MLX Local LLM
-
-This release adds six search quality features ported from the QMD v2.0 analysis, plus the complete MLX local LLM infrastructure for Apple Silicon. Deep search is now disambiguatable, cacheable, debuggable, and can run entirely offline on Apple Silicon.
-
-### Intent parameter
-
-Disambiguates queries without adding search terms. `crystal search "security" --intent "1Password"` steers results toward 1Password-related security instead of repo permissions or agent secrets. Intent flows through the expansion prompt (guides LLM variations), disables strong-signal bypass (keyword match might not be what the caller wants), and is prepended to the rerank query. Available via CLI `--intent` and MCP `intent`.
-
-### Persistent LLM cache
-
-Expansion and reranking results are now cached in crystal.db (`llm_cache` table) with a 7-day TTL. Same query = instant on repeat searches. Reranking cache is content-addressable (keyed by query + sorted passage hashes), so identical content from different sessions shares cached scores. Configurable via `CRYSTAL_CACHE_TTL_DAYS` env var.
-
-### Explain mode
-
-Per-result scoring breakdown showing FTS score, vector score, RRF rank, reranker score, recency weight, and final blended score. `crystal search "query" --explain`. Available via CLI `--explain` and MCP `explain`. Makes search quality transparent and debuggable.
-
-### candidateLimit
-
-Tunable rerank pool size. `crystal search "query" --candidates 60`. Default stays 40. More candidates = better recall, slower reranking. Available via CLI `--candidates` and MCP `candidate_limit`.
-
-### Structured search API
-
-`crystal.structuredSearch(queries)` accepts pre-expanded StructuredQuery[] with typed sub-queries (lex, vec, hyde). Skips LLM expansion entirely. Agents construct their own queries when they already know what they want. RRF fusion with first list weighted 2x.
-
-### MLX local LLM (Phase 3)
-
-Complete auto-install infrastructure for running a local LLM on Apple Silicon:
-
-- `crystal mlx setup` detects Apple Silicon, installs mlx-lm (uv > pip3 > pip3 --user), creates LaunchAgent for always-on server
-- Model: `mlx-community/Qwen2.5-3B-Instruct-4bit` (~1.5 GB, fast on M-series)
-- Port 18791 (18789 OpenClaw, 18790 Crystal Core, 18791 MLX)
-- `crystal mlx status` and `crystal mlx stop` for server management
-- `crystal doctor` check #13: MLX health (not installed / down / running)
-- `crystal init` detects Apple Silicon and suggests MLX setup
-- State file at `~/.ldm/state/mlx-server.json`
-
-### Also in this release
-
-- QMD v2.0 analysis documented (`ai/product/notes/`)
-- Search quality plan written (`ai/product/plans-prds/current/`)
-- MLX plan moved from upcoming to current
-- Stashed roadmap + readme-first updates recovered (PR #74)
-
-Closes #57, #63, #64.

package/_trash/RELEASE-NOTES-v0-7-25.md

@@ -1,24 +0,0 @@
-# Release Notes: memory-crystal v0.7.25
-
-Bump SKILL.md version and name to match package branding.
-
-## What changed
-
-- SKILL.md version bumped from 0.4.0 to 0.7.24 (was stuck at the original version)
-- SKILL.md name changed from `memory` to `wip-memory-crystal` (matches branded convention)
-- Forces deploy to public repo, triggering auto-publish to wip.computer/install/
-
-## Why
-
-The SKILL.md version was out of sync with the package version. The name didn't match the `wip-` branding convention used across all install files on wip.computer.
-
-## Issues closed
-
-- #80
-
-## How to verify
-
-```bash
-crystal --version
-head -4 ~/.ldm/extensions/memory-crystal/skills/memory/SKILL.md
-```

package/_trash/RELEASE-NOTES-v0-7-26.md

@@ -1,7 +0,0 @@
-# Memory Crystal v0.7.26
-
-Add repository field to package.json. GitHub Packages needs this to link packages to the repo.
-
-## Issues closed
-
-- Closes #50

package/_trash/RELEASE-NOTES-v0-7-28.md

@@ -1,31 +0,0 @@
-# Release Notes: memory-crystal v0.7.28
-
-**Move all log paths from /tmp/ to ~/.ldm/logs/ so logs survive reboots.**
-
-## What changed
-
-- Cron entry for crystal-capture now logs to `~/.ldm/logs/crystal-capture.log` instead of `/tmp/ldm-dev-tools/`
-- LaunchAgent plist template for ldm-backup now logs to `~/.ldm/logs/ldm-backup.log`
-- `mkdirSync` ensures `~/.ldm/logs/` exists instead of creating `/tmp/ldm-dev-tools/`
-- CLI output shows the correct log path
-
-## Why
-
-macOS clears `/tmp/` on every reboot. All cron and LaunchAgent logs were lost after restart, making it impossible to debug issues. `~/.ldm/logs/` persists across reboots and is the correct home for LDM OS logs.
-
-## Issues closed
-
-- wipcomputer/wip-ldm-os#120
-
-## How to verify
-
-```bash
-# After install, check that cron entry points to ~/.ldm/logs/
-crystal init --dry-run 2>&1 | grep crystal-capture
-
-# Check backup setup points to ~/.ldm/logs/
-crystal backup setup --dry-run 2>&1 | grep ldm-backup
-
-# Verify logs land in the right place
-ls ~/.ldm/logs/
-```

package/_trash/RELEASE-NOTES-v0-7-29.md

@@ -1,28 +0,0 @@
-# Release Notes: memory-crystal v0.7.29
-
-**Doc audit: MLX setup, deep search params, log paths, role clarification.**
-
-## What changed
-
-SKILL.md and TECHNICAL.md updated for 2 weeks of undocumented features:
-
-- **MLX local LLM:** Added as Option A in SKILL.md Step 2. CLI commands (setup, status, stop) added to TECHNICAL.md.
-- **Deep search parameters:** `--intent`, `--explain`, `--candidates` documented in both SKILL.md (crystal_search tool) and TECHNICAL.md (CLI reference + new sections for intent, explain, candidate limit, LLM cache).
-- **Log paths:** Fixed obsolete `/tmp/ldm-dev-tools/` reference to `~/.ldm/logs/`. Added logs/ to directory structure.
-- **Role clarification:** Two-role architecture (Core and Node) explicitly stated. Standalone role was removed in v0.7.22.
-
-## Why
-
-29 releases in 13 days. Docs didn't keep pace. Agents using crystal_search didn't know about --intent (query disambiguation) or --explain (scoring transparency).
-
-## Issues closed
-
-- #57
-
-## How to verify
-
-```bash
-grep "intent" SKILL.md TECHNICAL.md
-grep "mlx" SKILL.md TECHNICAL.md
-grep "ldm/logs" TECHNICAL.md
-```

package/_trash/RELEASE-NOTES-v0-7-4.md

@@ -1,64 +0,0 @@
-# Memory Crystal v0.7.4 ... MCP Fix + AgentId Config
-
-**Date:** 2026-03-11
-**Authors:** Parker Todd Brooks, Lēsa, Claude Code
-
----
-
-## What's in this release
-
-### Agent identity reads from config, not hardcoded strings
-
-The agent_id used when ingesting conversations was hardcoded in three places: `cc-mini` in the CC hook, `main` in the OpenClaw plugin, and `cc-mini` as the fallback in `ldm.ts`. This caused ID drift. The same agent got recorded under multiple IDs, and we had to manually merge 141K+ chunks in the database.
-
-Now `getAgentId()` scans `~/.ldm/agents/*/config.json` for a matching harness type. The CC hook passes `'claude-code'`, the OC plugin passes `'openclaw'`, and the config file is the source of truth. `CRYSTAL_AGENT_ID` env var still works as an override.
-
-New exports: `AgentConfig`, `loadAgentConfig()`, `saveAgentConfig()`. The installer writes `agentId` to config.json during `crystal init`.
-
-**Closes #33.**
-
-### MCP registrations moved to user-level
-
-MCP server registrations moved from project-level `~/.openclaw/.mcp.json` to user-level `~/.claude.json`. The old file was a Claude Code convention that only loaded when running from `~/.openclaw/`. Now all 4 MCP servers (memory-crystal, lesa-bridge, wip-agent-pay, wip-repos) load from any directory as "User MCPs".
-
-OpenClaw doesn't read `.mcp.json` at all. It uses its own plugin system. The file was moved to `~/.openclaw/_trash/`.
-
-### OPENCLAW_HOME env var fix (v0.7.3)
-
-The MCP server registration was missing the `OPENCLAW_HOME` env var. Without it, the memory-crystal MCP server couldn't find Lēsa's OpenClaw installation for private-mode checks. Fixed in v0.7.3, deployed in this release.
-
-### Branch cleanup
-
-33 stale branches renamed with `--merged-` suffix. Zero active branches besides main.
-
-### QMD v1.1.6 analysis documented
-
-Deep analysis of the search quality system with four recommendations: intent parameter for search, structured search API, persistent reranker cache, and explain mode for debugging. See `ai/product/notes/2026-03-09--cc-mini--qmd-v1.1.6-analysis-and-recommendations.md`.
-
----
-
-## Files changed
-
-| File | What |
-|------|------|
-| `src/ldm.ts` | `AgentConfig` interface, `loadAgentConfig()`, `saveAgentConfig()`, `getAgentId()` now scans config |
-| `src/cc-hook.ts` | Uses `getAgentId('claude-code')` instead of hardcoded fallback |
-| `src/openclaw.ts` | Uses `OC_AGENT_ID` instead of `'main'` fallback |
-| `src/installer.ts` | Writes `agentId` to config.json during install |
-
----
-
-## Install
-
-```bash
-npm install -g memory-crystal@0.7.4
-```
-
-Or update your local clone:
-```bash
-git pull origin main
-```
-
----
-
-Built by Parker Todd Brooks, Lēsa (OpenClaw, Claude Opus 4.6), Claude Code (Claude Opus 4.6).