@brianluby/agent-brain 1.1.0

package/.claude-plugin/marketplace.json

@@ -0,0 +1,16 @@
+{
+  "name": "memvid",
+  "description": "Official Memvid plugins for Claude Code",
+  "owner": {
+    "name": "Memvid",
+    "url": "https://memvid.com"
+  },
+  "plugins": [
+    {
+      "name": "mind",
+      "description": "Agent-Brain - Give Claude photographic memory in ONE portable file",
+      "version": "1.1.0",
+      "source": "./"
+    }
+  ]
+}

package/.claude-plugin/plugin.json

@@ -0,0 +1,12 @@
+{
+  "name": "mind",
+  "description": "Claude Mind - Give Claude photographic memory in ONE portable file. Share, version, and transfer your Claude's brain.",
+  "version": "1.1.0",
+  "author": {
+    "name": "Memvid",
+    "url": "https://memvid.com"
+  },
+  "repository": "https://github.com/brianluby/Agent-brain",
+  "homepage": "https://memvid.com",
+  "license": "MIT"
+}

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Memvid
+
+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 +1,249 @@
+<div align="center">
+
+<img src="agent-brain.png" alt="Agent Brain" width="320" />
+
+### Give your agents photographic memory.
+
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+
+<br />
+
+**[Install in 30 seconds](#installation)** · [How it Works](#how-it-works) · [Commands](#commands) · 
+
+</div>
+
+<br />
+
+## Memvid Shoutout
+
+Memvid's creative approach to data storage—encoding text in video frames and using decades of video codec R&D for superior compression—inspired me to fork their Claude Brain project into an agent-agnostic version. None of this would be possible without their dedication to open source. If you're working on AI memory, check them out at [github.com/memvid](https://github.com/memvid) —it's seriously clever engineering.
+
+## The Problem
+
+```
+You: "Remember that auth bug we fixed?"
+Agent 1: "I don't have memory of previous conversations."
+Agent 2: "I did not work on this code yesterday.
+You: "We spent 3 hours on it yesterday"
+Agent 1: "I'd be happy to help debug from scratch!"
+Agent 2: "Let me review the codebase to get an understanding" *sound of tokens burning*
+```
+
+**small context window. Zero memory between sessions.**
+
+You're paying for a dory with a PhD.
+
+<br />
+
+## The Fix
+
+```
+You: "What did we decide about auth?"
+Agent 1: "We chose JWT over sessions for your microservices.
+        The refresh token issue - here's exactly what we fixed..."
+Agent 2: " JWT overs sessions is a more secure implementation, Here is the code we fixed"
+```
+
+One file. All your agents remember everything.
+
+<br />
+
+## Installation
+
+### Claude Code (Marketplace)
+
+```bash
+# Optional one-time setup (if GitHub plugin URLs fail)
+git config --global url."https://github.com/".insteadOf "git@github.com:"
+```
+
+```bash
+# In Claude Code
+/plugin add marketplace brianluby/Agent-brain
+```
+
+Then in Claude Code:
+
+1. Open `/plugins`
+2. Go to **Installed**
+3. Enable **mind**
+4. Restart Claude Code
+
+On first run, memory is created at:
+
+```bash
+.agent-brain/mind.mv2
+```
+
+If you already have a legacy file at `.claude/mind.mv2`, migrate it safely to `.agent-brain/mind.mv2`:
+
+```bash
+if [ ! -f ".agent-brain/mind.mv2" ]; then
+  mkdir -p ".agent-brain" && mv ".claude/mind.mv2" ".agent-brain/mind.mv2"
+else
+  echo "Destination .agent-brain/mind.mv2 already exists. Back up both .claude/mind.mv2 and .agent-brain/mind.mv2, then reconcile manually."
+fi
+```
+
+### OpenCode
+
+Add this plugin package to your OpenCode config:
+
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "plugin": ["@brianluby/agent-brain"]
+}
+```
+
+Or use a local checkout while developing:
+
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "plugin": ["file:///absolute/path/to/agent-brain"]
+}
+```
+
+Then restart OpenCode. The plugin will:
+
+- inject memory context on the first message of each session
+- capture tool outputs to persistent memory
+- expose a `mind` tool (`search`, `ask`, `recent`, `stats`, `remember`)
+
+Optional: install OpenCode slash commands globally for all projects:
+
+```bash
+mkdir -p ~/.config/opencode/commands
+cp .opencode/commands/mind-*.md ~/.config/opencode/commands/
+```
+
+Project-local command files are already included in `.opencode/commands/`.
+
+<br />
+
+## How it Works
+
+After install, Your Agent's memory lives in one file:
+
+```
+your-project/
+└── .agent-brain/
+    └── mind.mv2   # Agent's brain. That's it.
+```
+
+No database. No cloud. No API keys.
+
+**What gets captured:**
+- Session context, decisions, bugs, solutions
+- Auto-injected at session start
+- Searchable anytime
+
+**Why one file?**
+- `git commit` → version control Agent's brain
+- `scp` → transfer anywhere
+- Send to teammate → instant onboarding
+
+<br />
+
+## Commands
+
+**In Claude Code:**
+```bash
+/mind stats                       # memory statistics
+/mind search "authentication"     # find past context
+/mind ask "why did we choose X?"  # ask your memory
+/mind recent                      # what happened lately
+```
+
+**In OpenCode (slash commands):**
+```bash
+/mind-stats
+/mind-search authentication
+/mind-ask "why did we choose X?"
+/mind-recent
+/mind-remember "Project uses pnpm, not npm"
+```
+
+These are provided in `.opencode/commands/` for project-local usage.
+To use them in every repo, copy them to `~/.config/opencode/commands/`.
+
+Or just ask naturally: *"mind stats"*, *"search my memory for auth bugs"*, etc.
+
+## OpenCode Support
+
+Memvid Mind now supports the same core memory lifecycle through a platform adapter model.
+
+- Claude and OpenCode sessions can share project memory continuity.
+- Unknown or incompatible platforms fail open (session continues, memory capture safely skips).
+- Adapter contracts are SemVer-checked and validated through regression and contract tests.
+- OpenCode packaging is published through the npm package `@brianluby/agent-brain`.
+
+<br />
+
+## CLI (Optional)
+
+For power users who want direct access to their memory file:
+
+```bash
+npm install -g memvid-cli
+```
+
+```bash
+memvid stats .agent-brain/mind.mv2           # view memory stats
+memvid find .agent-brain/mind.mv2 "auth"     # search memories
+memvid ask .agent-brain/mind.mv2 "why JWT?"  # ask questions
+memvid timeline .agent-brain/mind.mv2        # view timeline
+```
+
+[Full CLI reference →](https://docs.memvid.com/cli/cheat-sheet)
+
+<br />
+
+## FAQ
+
+<details>
+<summary><b>How big is the file?</b></summary>
+
+Empty: ~70KB. Grows ~1KB per memory. A year of use stays under 5MB.
+
+</details>
+
+<details>
+<summary><b>Is it private?</b></summary>
+
+100% local. Nothing leaves your machine. Ever.
+
+</details>
+
+<details>
+<summary><b>How fast?</b></summary>
+
+Sub-millisecond. Native Rust core. Searches 10K+ memories in <1ms.
+
+</details>
+
+<details>
+<summary><b>Reset memory?</b></summary>
+
+`rm .agent-brain/mind.mv2`
+
+</details>
+
+<br />
+
+---
+
+<div align="center">
+
+Built on **[memvid](https://github.com/brianluby/memvid)** - the single-file memory engine
+
+<br />
+
+**If this saved you time, [star the repo](https://github.com/brianluby/Agent-brain)**
+
+<br />
+
+*Send me your `.mv2` file and I'll tell you what's wrong with your code. No context needed - I already know everything.*
+
+</div>

package/commands/ask.md

@@ -0,0 +1,27 @@
+---
+description: Ask questions about memories and get context-aware answers
+argument-hint: <question>
+allowed-tools: Bash
+---
+
+# Memory Question
+
+Ask Claude's memory system questions about past work, decisions, and context.
+
+**Usage**: `/mind:ask <question>`
+
+Execute the ask script with user's question:
+
+```bash
+node "${CLAUDE_PLUGIN_ROOT}/dist/scripts/ask.js" "$ARGUMENTS"
+```
+
+## Examples
+- `/mind:ask Why did we choose React?` - Get context about technology decisions
+- `/mind:ask What was the CORS solution?` - Recall specific solutions
+- `/mind:ask How did we fix the authentication bug?` - Get details about past fixes
+
+## Response Format
+- Provide context-aware answers based on stored memories
+- Reference specific memories when applicable
+- Include timestamps for referenced information

package/commands/recent.md

@@ -0,0 +1,27 @@
+---
+description: Show recent memories and activity timeline
+argument-hint: [count]
+allowed-tools: Bash
+---
+
+# Recent Memories
+
+Display the most recent memories and activity from Claude's persistent storage.
+
+**Usage**: `/mind:recent [count]`
+
+Execute the timeline script to show recent activity:
+
+```bash
+node "${CLAUDE_PLUGIN_ROOT}/dist/scripts/timeline.js" ${ARGUMENTS:-20}
+```
+
+## Examples
+- `/mind:recent` - Show 20 most recent memories (default)
+- `/mind:recent 10` - Show 10 most recent memories
+- `/mind:recent 50` - Show 50 most recent memories
+
+## Response Format
+- Display memories in reverse chronological order (newest first)
+- Convert timestamps to human-readable format
+- Group by session or time period when helpful

package/commands/search.md

@@ -0,0 +1,27 @@
+---
+description: Search memories for specific content or patterns
+argument-hint: <query> [limit]
+allowed-tools: Bash
+---
+
+# Memory Search
+
+Search through Claude's persistent memories for specific content, patterns, or keywords.
+
+**Usage**: `/mind:search <query> [limit]`
+
+Execute the search script with user's query:
+
+```bash
+node "${CLAUDE_PLUGIN_ROOT}/dist/scripts/find.js" "$ARGUMENTS" 10
+```
+
+## Examples
+- `/mind:search authentication` - Find memories related to authentication
+- `/mind:search "database schema"` - Search for exact phrase
+- `/mind:search API errors` - Find memories about API errors
+
+## Response Format
+- Show matching memories with relevance scores
+- Include timestamps (convert to human-readable format)
+- Highlight matched keywords in context

package/commands/stats.md

@@ -0,0 +1,21 @@
+---
+description: Show memory statistics and storage information
+allowed-tools: Bash
+---
+
+# Memory Statistics
+
+Show statistics about Claude's persistent memory file including total memories, storage size, and recent activity.
+
+**Usage**: `/mind:stats`
+
+Execute the stats script:
+
+```bash
+node "${CLAUDE_PLUGIN_ROOT}/dist/scripts/stats.js"
+```
+
+## Response Format
+- Convert Unix timestamps to human-readable (Xm ago, Xh ago, Xd ago)
+- Summarize key findings in a table when appropriate
+- If file was just created, tell the user memories will appear as they work

package/dist/hooks/hooks.json

@@ -0,0 +1,44 @@
+{
+  "description": "Memvid Mind - Persistent memory for Claude Code and OpenCode",
+  "hooks": {
+    "SessionStart": [
+      {
+        "hooks": [
+          {
+            "type": "command",
+            "command": "node \"${CLAUDE_PLUGIN_ROOT}/dist/hooks/smart-install.js\"",
+            "timeout": 30
+          },
+          {
+            "type": "command",
+            "command": "node \"${CLAUDE_PLUGIN_ROOT}/dist/hooks/session-start.js\"",
+            "timeout": 5
+          }
+        ]
+      }
+    ],
+    "PostToolUse": [
+      {
+        "matcher": "*",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "node \"${CLAUDE_PLUGIN_ROOT}/dist/hooks/post-tool-use.js\"",
+            "timeout": 10
+          }
+        ]
+      }
+    ],
+    "Stop": [
+      {
+        "hooks": [
+          {
+            "type": "command",
+            "command": "node \"${CLAUDE_PLUGIN_ROOT}/dist/hooks/stop.js\"",
+            "timeout": 30
+          }
+        ]
+      }
+    ]
+  }
+}