ATLAS MCP Server
An integrated advisory system that transforms Claude into a personal career coach, life strategist, and business consultant with persistent memory, accountability tracking, and multi-lens analysis.
README
ATLAS — Adaptive Total Life Advisory System
Your AI-powered career coach, life strategist, and business consultant.
ATLAS is a Claude Code plugin that transforms Claude into a retained personal advisor — one that remembers your context, tracks your commitments, flags contradictions, and holds you accountable.
┌─────────────────────────────┐
│ ATLAS MCP │
│ Advisory Intelligence │
└──────────┬──────────────────┘
│
┌────────────────┼────────────────┐
│ │ │
┌────────▼───────┐ ┌─────▼──────┐ ┌───────▼────────┐
│ Therapist │ │ Business │ │ Functional │
│ Lens │ │ Coach │ │ Advisor │
│ │ │ Lens │ │ Lens │
│ • Patterns │ │ • Strategy │ │ • Genetics │
│ • Behaviors │ │ • Markets │ │ • Energy │
│ • Psychology │ │ • Accounta- │ │ • Health │
│ • Blind spots │ │ bility │ │ • Optimization │
└────────┬───────┘ └─────┬──────┘ └───────┬────────┘
│ │ │
└────────────────┼────────────────┘
│
┌──────────▼──────────────────┐
│ Integrated Advisory Output │
│ Grounded in verified data │
└─────────────────────────────┘
Install as Claude Code Plugin (Recommended)
Step 1 — Add the marketplace
In Claude Code, run:
/config
Then add to your settings.json:
{
"extraKnownMarketplaces": {
"atlas": {
"source": {
"source": "github",
"repo": "Ak47dev-biz/atlas-mcp-server"
}
}
},
"enabledPlugins": {
"atlas@atlas": true
}
}
Step 2 — Restart Claude Code
The plugin loads on startup. Once active, you get four slash commands:
| Command | Description |
|---|---|
/atlas |
Start an advisory session |
/atlas-onboard |
First-time setup — create your client profile |
/atlas-review |
Weekly accountability and progress review |
/atlas-opportunity |
Deep-dive multi-lens analysis of a specific opportunity |
Step 3 — Get started
/atlas-onboard
This walks you through a guided intake conversation to build your profile.
Alternative: Standalone MCP Server
If you prefer to use ATLAS as a standalone MCP server without the plugin system:
Claude Code
claude mcp add atlas node /absolute/path/to/atlas-mcp-server/dist/index.js
Remove when done:
claude mcp remove atlas
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):
{
"mcpServers": {
"atlas": {
"command": "node",
"args": ["/absolute/path/to/atlas-mcp-server/dist/index.js"],
"env": {
"ATLAS_DATA_DIR": "/path/to/your/atlas/data"
}
}
}
}
Build from Source
git clone https://github.com/Ak47dev-biz/atlas-mcp-server.git
cd atlas-mcp-server
npm install
npm run build
Features
- Three Advisory Lenses — Therapist (behavioral patterns), Business Coach (strategy & accountability), Functional Advisor (health & optimization)
- Persistent Memory — All data stored locally as JSON files. Nothing leaves your machine.
- Slash Commands —
/atlas,/atlas-onboard,/atlas-review,/atlas-opportunity - Accountability Tracking — Commitments with deadlines and automatic overdue detection
- Advice Consistency — Every piece of advice is logged so Claude can check past recommendations
- Anti-Hallucination — System prompt enforces that Claude only uses verified facts, never invents details
- Document Ingestion — Feed in resumes, LinkedIn profiles, chat histories, genetic data, and health markers
- Structured Frameworks — Career decision matrices, business decision frameworks, opportunity analysis
- Session Continuity — Sessions are logged so Claude picks up where you left off
Slash Commands
/atlas — Start a Session
Opens a full advisory session. ATLAS loads your profile, checks pending commitments, and greets you with context. Use this for career discussions, decision-making, goal-setting, or any advisory need.
/atlas-onboard — First-Time Setup
Guided onboarding conversation that builds your client profile across all three lenses. Covers career background, goals, decision-making patterns, and optional health/genetic data.
/atlas-review — Weekly Review
Structured accountability review:
- Commitment status (completed, pending, overdue)
- Goal progress and milestone tracking
- Pattern detection across all three lenses
- Forward planning with new commitments
/atlas-opportunity — Opportunity Analysis
Deep-dive analysis of a specific opportunity (job offer, business deal, partnership, career move) using all three advisory lenses with a structured recommendation.
Tools Reference
Profile Management
| Tool | Description |
|---|---|
atlas_onboard_client |
Create a new client profile |
atlas_get_profile |
Retrieve full profile across all lenses |
atlas_update_profile |
Update specific profile sections |
Document Ingestion
| Tool | Description |
|---|---|
atlas_ingest_resume |
Parse and store career facts from resume text |
atlas_ingest_linkedin |
Store LinkedIn profile data |
atlas_ingest_chat_history |
Analyze past conversations for behavioral patterns |
atlas_ingest_genetic_data |
Store SNP/genetic data (23andMe, Ancestry, etc.) |
atlas_ingest_health_data |
Store health markers, labs, sleep, nutrition |
Structured Data
| Tool | Description |
|---|---|
atlas_add_career_fact |
Add a single verified career entry |
atlas_add_pattern |
Record an observed behavioral pattern |
atlas_add_goal |
Track a new goal with milestones |
atlas_update_goal |
Update goal status or complete milestones |
Accountability
| Tool | Description |
|---|---|
atlas_track_commitment |
Log a commitment with a deadline |
atlas_complete_commitment |
Mark a commitment as completed |
atlas_review_accountability |
Full review with automatic overdue detection |
Advisory
| Tool | Description |
|---|---|
atlas_log_session |
Log an advisory session for continuity |
atlas_search_advice_history |
Search past advice for consistency |
atlas_analyze_opportunity |
Structured multi-lens opportunity analysis |
Resources
Resources provide read-only access to ATLAS data:
| URI | Description |
|---|---|
atlas://profile/summary |
Client profile overview |
atlas://profile/facts |
All verified career facts |
atlas://profile/patterns |
Observed behavioral patterns |
atlas://profile/goals |
Active goals and milestones |
atlas://profile/accountability |
Commitments and tracking |
atlas://profile/functional |
Genetic and health data |
atlas://history/sessions |
Past session logs |
atlas://history/advice |
Past advice for consistency |
Data Storage
All data is stored locally as JSON files. No external services, no databases, no cloud.
~/.atlas/data/ # Default (override with ATLAS_DATA_DIR)
├── profile.json # Core client profile
├── facts/
│ ├── career.json # Verified career facts
│ └── achievements.json # Notable achievements
├── patterns/
│ ├── behavioral.json # Observed patterns
│ └── themes.json # Recurring themes
├── goals/
│ └── active.json # Goals with milestones
├── accountability/
│ ├── commitments.json # Tracked commitments
│ └── advice_log.json # Past advice for consistency
├── functional/
│ ├── genetic.json # SNP/genetic data
│ └── health.json # Health markers
├── documents/ # Raw ingested documents
│ ├── resume.json
│ ├── linkedin.json
│ └── chat_*.json
└── sessions/
└── log.json # Session history
Your data is yours. Back it up, version control it, or delete it anytime.
Philosophy
ATLAS exists because the best advice comes from advisors who know you deeply — your patterns, your goals, your contradictions, and your context. Most people can't afford a team of a therapist, business coach, and health optimizer working together. ATLAS bridges that gap by giving Claude the memory and structure to serve as that integrated advisory team.
Principles:
- Privacy first — Your data never leaves your machine. Period.
- Facts over assumptions — ATLAS only works with verified information.
- Accountability over comfort — Good advisors tell you what you need to hear.
- Integration over isolation — Career, psychology, and health are interconnected.
- Tools over magic — The MCP server is a data layer. Claude provides the intelligence.
Contributing
Contributions welcome! Areas where help is appreciated:
- Additional ingestion tools — support for more data formats
- Visualization — dashboards for goals, patterns, accountability
- Export — structured reports in various formats
- Testing — test coverage for all tools
- Documentation — guides for specific use cases
- New skills — additional slash commands for specific workflows
Please open an issue before starting work on major features.
Roadmap
- [x] Core MCP server with tools, resources, and prompts
- [x] Local JSON storage
- [x] Three advisory lenses
- [x] Accountability tracking with overdue detection
- [x] Advice consistency logging
- [x] Claude Code plugin with slash commands
- [ ] GitHub Actions CI pipeline
- [ ] Comprehensive test suite
- [ ] Data export and reporting
- [ ] Web-based dashboard
- [ ] Multi-client support
- [ ] Encrypted data storage
- [ ] Plugin system for custom lenses
License
MIT — see LICENSE.
Disclaimer
ATLAS is an AI advisory tool, not a replacement for licensed professionals. It does not provide medical, legal, or financial advice. Always consult qualified professionals for decisions in those domains. ATLAS provides strategic context, pattern recognition, and accountability — the thinking is yours.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.