Power your agent
การติดตั้งจากผู้เผยแพร่ ไม่ใช่เอนด์พอยต์ของ PlugMCP
URL แบบรีโมตหรือคำสั่งบนเครื่องของคุณ? การติดตั้งสองแบบต่างกันอย่างไร
รันบน PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
เปิดจากผู้เผยแพร่
Works with Cursor, Claude Code, and any MCP-compatible AI assistant.
Solve AI assistant memory loss and context confusion. Cursor-Cortex gives your AI the structured thinking and long-term memory it needs to actually help you build software.
Structured memory system for AI assistants. Eliminates context loss with branch notes, tacit knowledge, and project context. Local MCP integration for Cursor IDE.
LLM-based AI assistants are amazing at generating code and explaining concepts, but they struggle with:
Result: You spend more time explaining context than actually solving problems.
Cursor-Cortex provides your AI with structured external memory:
Your AI always knows the project goals, architecture decisions, and current focus.
Continuous development log so your AI can pick up exactly where you left off.
Company-specific solutions, patterns, and hard-won insights that LLMs don't know.
Structured thinking frameworks that keep AI (and you) organized and thorough.
You: "Help me continue the authentication work"
AI: "I need more context. What authentication system? What issues were you facing?"
You: "Ugh, let me explain everything again..."
You: "Help me continue the authentication work"
AI: *reads context + branch notes + knowledge base*
"I see you're implementing JWT refresh tokens and hit CORS issues yesterday.
Based on our company knowledge doc about API security, let's check the middleware..."
Turn your AI from a smart intern into an experienced teammate who remembers everything, follows structured thinking, and has access to institutional knowledge.
npm install --legacy-peer-deps
Note: The
--legacy-peer-depsflag is required due to TensorFlow.js peer dependency conflicts between v3.x and v4.x versions.
pwd
# Copy the output - you'll need it for configuration
For AI-powered semantic search capabilities:
System Requirements:
Setup Steps:
# Step 1: Generate embeddings for all knowledge files
node generate-all-embeddings-cpu.js
# This creates vector embeddings for:
# - All tacit knowledge documents
# - All branch notes
# - All context files
# Duration varies by document count and system specs
# Step 2 (Optional): Force regenerate if content changes
node generate-all-embeddings-cpu.js --force
# Step 3 (Optional): View progress during generation
node generate-all-embeddings-cpu.js --verbose
Why This Matters:
~/.cursor-cortex/embeddings/Create or edit ~/.cursor/mcp.json:
"cursor-cortex": {
"command": "node",
"args": ["/YOUR/FULL/PATH/TO/cursor-cortex/index.js"],
"tools": [
"update_branch_note",
"add_commit_separator",
"read_branch_notes",
"update_context_file",
"read_project_context",
"read_branch_context",
"list_context_files",
"generate_commit_message",
"generate_jira_comment",
"create_tacit_knowledge",
"read_tacit_knowledge",
"search_branch_notes",
"generate_embeddings",
"create_completion_checklist",
"read_checklist",
"update_checklist",
"sign_off_checklist",
"archive_branch_note",
"clear_branch_note",
"list_all_branch_notes",
"enhanced_branch_survey",
"comprehensive_knowledge_search",
"construct_project_narrative",
"timeline_reconstruction",
"context_sync_guidance",
"analyze_documentation_gaps",
"migrate_context_files",
"request_critical_thinking_space",
"check_critical_thinking_status",
"request_thinking_guidance",
"add_perspective",
"request_synthesis_space",
"request_synthesis_step_guidance",
"complete_synthesis_step",
"generate_context_zip",
"unpack_context"
],
"stdio": true,
"env": {
"DEBUG": "true"
}
},
Replace /YOUR/FULL/PATH/TO/cursor-cortex/ with the output from step 2
Note: This configuration is for Cursor-Cortex v1.2.0+. If you're using an older version, some tools may not be available. Check your
package.jsonversion and consider updating.
Close and reopen Cursor completely.
Create or edit .claude/settings.json in your project root (or ~/.claude/settings.json globally):
{
"mcpServers": {
"cursor-cortex": {
"command": "node",
"args": ["/YOUR/FULL/PATH/TO/cursor-cortex/index.js"]
}
}
}
Replace /YOUR/FULL/PATH/TO/cursor-cortex/ with the output from step 2, then restart Claude Code.
Storage directory created at /Users/yourname/.cursor-cortexIn any Cursor chat, try:
"Update my branch notes: Testing Cursor-Cortex setup"
If it works, you'll see: Successfully updated branch note with: "Testing Cursor-Cortex setup"
Once everything is set up, explore your knowledge base as an interactive network graph:
npm run explorer
Then open http://localhost:4000 in your browser.
Features:
❌ "Tool not found" error
❌ MCP configuration not working
❌ Vector search not working
npm install to ensure TensorFlow.js is installedNeed Help? Create an issue with your error message for quick support.
As you code, tell your AI assistant:
"Update branch notes: Fixed the login validation bug by adding proper error handling
for expired tokens. Had to modify the JWT middleware."
When you return to work:
"Show me my uncommitted work"
When ready to commit:
"Generate a commit message"
AI reads all your notes and creates: "Fix JWT token validation and error handling"
Cursor-Cortex organizes your knowledge into:
All stored in ~/.cursor-cortex/ and accessible through AI.
Once you're comfortable with basics, explore these advanced capabilities:
"Search all my projects for JWT authentication solutions"
node setup-hooks.js # Auto-separates pre/post commit work
Track requirements, testing, and sign-offs:
"Create a completion checklist for the user auth feature"
Create reusable solution documents:
"Create tacit knowledge document about our JWT implementation approach"
Analyze and synthesize project knowledge:
"Survey all my branch documentation and show completeness scores"
"Construct a project narrative for the user authentication system"
"Show me a timeline of all development across projects"
"Analyze this folder structure for documentation gaps"
update_branch_note - Add entries to your development logread_branch_notes - Default is the full file (including COMMIT separators); mode=uncommitted or uncommittedOnly=true for work since the last commit; optional beforeDate, afterDate, or commitHash to filteradd_commit_separator - Mark commit boundaries (auto via git hooks)generate_commit_message - AI creates commits from your notesupdate_context_file - Document project goals and decisionsread_project_context - Read branch-agnostic project context onlyread_branch_context - Read branch-specific context onlylist_context_files - Browse all project docs with cross-project warningscreate_tacit_knowledge - Document solutions for future referenceread_tacit_knowledge - Search knowledge across all projects with tags and content searchsearch_branch_notes - Search branch notes across projects with semantic searchgenerate_embeddings - Generate or regenerate vector embeddings for semantic searchcomprehensive_knowledge_search - Global semantic search across ALL Cursor-Cortex knowledgecreate_completion_checklist - Track requirements and deliverablesread_checklist - View progressupdate_checklist - Mark items completed (manual or auto-detect)sign_off_checklist - Formal approval with signaturesarchive_branch_note - Archive completed workclear_branch_note - Reset branch documentationmigrate_context_files - Migrate existing context files to Smart Hybrid Context Systemlist_all_branch_notes - View all branch notes across all projects with priority orderingenhanced_branch_survey - Comprehensive analysis of documentation with completeness scoring and relationship mappingconstruct_project_narrative - Weave scattered technical details into coherent production storiestimeline_reconstruction - Extract chronological timeline data from branch notes and commit separatorscontext_sync_guidance - Get contextual guidance for syncing project contexts based on timeline dataanalyze_documentation_gaps - Analyze folder structure to identify documentation needs and auto-create checklistsrequest_critical_thinking_space - Create systematic analysis workspace using Six Thinking Hats methodologycheck_critical_thinking_status - Check completion status of critical thinking analysisrequest_thinking_guidance - Get guidance for specific Six Thinking Hats perspectivesadd_perspective - Add specific perspective analysis to critical thinking workspacerequest_synthesis_space - Load perspectives into synthesis workspace for integrationrequest_synthesis_step_guidance - Get detailed guidance for synthesis stepscomplete_synthesis_step - Complete specific synthesis step and update processgenerate_context_zip - Package knowledge into portable ZIP archive for sharing between users/environmentsunpack_context - Import shared context package with safe conflict resolutionAutomatic commit separators require the MCP server to be running:
node setup-hooks.js # Install git hooks
Direct access without AI:
node cursor-cortex-cli.js # Interactive menu
Found a bug or have an idea? Open an issue or submit a pull request — contributions are welcome!
Author: Manuel Flores-Ramirez
This project is licensed under the Parachute Public License (PPL) v1.0.
The PPL is an ethical license that includes restrictions against use by government, military, surveillance, and other harmful entities. For full details and rationale, see parachute.pub.
© 2025 Manuel Flores-Ramirez.
Structured memory system for AI assistants. Eliminates context loss with branch notes, tacit knowledge, and project context. Local MCP integration for Cursor IDE.
เครื่องมือ
ยังไม่มีรายการเครื่องมือ เซิร์ฟเวอร์แบบรีโมตจะได้รายการหลังการตรวจสอบสถานะครั้งแรก
FAQ
คัดลอกคอนฟิก stdio สำหรับ Cursor, Claude Desktop, Claude Code หรือ VS Code ได้ในหน้านี้ คอนฟิกนี้ใช้การติดตั้งของผู้เผยแพร่ ไม่ใช่พร็อกซีของ PlugMCP
ระเบียนนี้ไม่ได้ระบุตัวแปรสภาพแวดล้อมที่จำเป็น
Cursor Cortex ทำงานแบบโลคัลผ่าน stdio
ตรวจสอบล่าสุดเมื่อ 2026-10-28 Structured memory system for AI assistants. Eliminates context loss with branch notes, tacit knowledge, and project context. Local MCP integration for Cursor IDE.
