Pickaxe AI Agent MCP

Manage your pickaxe.co AI agents, knowledge bases, users, and analytics directly through natural language.

An MCP (Model Context Protocol) server that connects AI assistants like Claude to thePickaxeplatform. Manage your AI agents, knowledge bases, users, and analytics directly through natural language.

- Node.js 18+
- APickaxeaccount with API access
- Your Pickaxe Studio API key(s)

git clone https://github.com/aplaceforallmystuff/mcp-pickaxe.git cd mcp-pickaxe npm install npm run build

- Log in toPickaxe Studio
- Navigate to Settings > API
- Copy your Studio API key (starts withstudio-)

{ "mcpServers": { "pickaxe": { "command": "node", "args": ["/path/to/mcp-pickaxe/dist/index.js"], "env": { "PICKAXE_STUDIO_MAIN": "studio-your-api-key-here" } } } }

{ "mcpServers": { "pickaxe": { "command": "node", "args": ["/path/to/mcp-pickaxe/dist/index.js"], "env": { "PICKAXE_STUDIO_MAIN": "studio-your-api-key-here" } } } }

To work with multiple Pickaxe studios, add multiple environment variables:

{ "env": { "PICKAXE_STUDIO_PRODUCTION": "studio-xxx-xxx-xxx", "PICKAXE_STUDIO_STAGING": "studio-yyy-yyy-yyy", "PICKAXE_STUDIO_DEV": "studio-zzz-zzz-zzz", "PICKAXE_DEFAULT_STUDIO": "PRODUCTION" } }

Then specify which studio to use in your requests:

- If you setPICKAXE_DEFAULT_STUDIO, that studio is used when none is specified
- If only one studio is configured, it's used automatically
- Otherwise, passstudio="STAGING"(or similar) to any tool

These are real workflows built with mcp-pickaxe in production environments.

Scenario:Detect prompt injection attempts across 29+ AI agents in real-time.

Implementation:An n8n workflow pollschat_historyhourly for all agents, runs messages against injection detection patterns (stored in Notion), and routes alerts by severity:

- HIGH/CRITICAL → Telegram alert + Notion log
- LOW/MEDIUM → Notion log only

n8n Schedule (hourly) → Fetch patterns from Notion → Loop through 29 pickaxe IDs → Fetch chat_history for each → Detect injections (regex patterns) → Route by severity → Alert/Log

Result:Real-time security monitoring across an entire studio with dynamic pattern management and severity-based alerting.

2. Knowledge Base Auto-Research Pipeline

Scenario:Automatically fact-check and maintain 31+ knowledge base articles.

Implementation:An n8n workflow queries KB articles from Notion, extracts key claims, fact-checks via Perplexity API, classifies changes by risk level, and routes to auto-update or human review.

Daily Schedule (2am) → Query KB articles from Notion → Filter by day (hash-based, ~1/7th daily) → Extract key claims → Perplexity fact-check → Classify: none/low/major risk → Route: auto-update or create review task

Tools used:doc_list,doc_get,doc_create,doc_connect

Result:KB content stays current with automated fact-checking and human-in-the-loop for major changes.

Scenario:Quarterly review of a training studio to identify KB gaps and user pain points.

1. "Fetch chat history from my training agents" 2. "Analyze: which questions got unclear or uncertain responses?" 3. "List all KB documents - which topics are missing?" 4. "Check user stats - who's most active, who's churning?" 5. "Create KB documents addressing the top 3 gaps" 6. "Connect new documents to the relevant agents"

Tools used:chat_history,doc_list,doc_create,doc_connect,user_list

Result:Data-driven KB improvements based on actual user conversations rather than guesswork.

Scenario:Managing multiple Pickaxe studios from a single Claude session.

{ "env": { "PICKAXE_STUDIO_PRODUCTION": "studio-xxx", "PICKAXE_STUDIO_STAGING": "studio-yyy", "PICKAXE_STUDIO_DEV": "studio-zzz", "PICKAXE_DEFAULT_STUDIO": "PRODUCTION" } }

1. "List users in PRODUCTION - how many signups this month?" 2. "Switch to STAGING - list products" 3. "Compare KB document counts across all studios" 4. "Find which studio has the most chat activity"

Tools used:studios_list,user_list,doc_list,products_list

Result:Cross-studio visibility without switching contexts or API keys manually.

Scenario:Review what your agents remember about users for personalization and privacy compliance.

1. "List all memory schemas defined in the studio" 2. "Get memories for user@example.com" 3. "What does the system know about this user's situation?" 4. "Which memory fields are most populated across users?"

User: maria.example@email.com Nickname: "Cautious Educator from Madrid" Summary: "Teaching [language] for [platform] at low hourly rate, considering self-employment status due to uncertain income" Memories: 1 stored

Tools used:memory_list,memory_get_user,user_list

Result:Visibility into personalization data for both product improvement and GDPR compliance.

Once configured, you can interact with Pickaxe through natural language:

"Show me the last 20 conversations from my support agent"

"What questions are users asking that my agent can't answer?"

"Create a new document called 'FAQ' with this content: [your content]"

"Connect the FAQ document to my customer support agent"

"List all documents in my knowledge base"

"Show me all users and their usage stats"

"Create a new user with emailuser@example.comand give them access to the Pro product"

"Send invitations to these emails: [list of emails]"

"List all users in my staging studio"

"Compare the documents between production and staging"

- studios_list- List all configured studios and the current default

- chat_history- Fetch conversation history for an agent

- Parameters:pickaxeId,skip,limit,format("messages" or "raw"),studio

- doc_create- Create document from content or URL
- doc_list- List all documents (with pagination)
- doc_get- Get a specific document
- doc_delete- Delete a document
- doc_connect- Link document to an agent
- doc_disconnect- Unlink document from an agent

- user_list- List all users with access and usage info
- user_get- Get a specific user by email
- user_create- Create a new user
- user_update- Update user details, products, or usage
- user_delete- Delete a user
- user_invite- Send email invitations

- products_list- List available products/bundles

- memory_list- List memory schemas
- memory_get_user- Get collected memories for a user

# Run in development mode (auto-reloads) npm run dev # Build for production npm run build # Run the built version npm start

"No Pickaxe studios configured"

Ensure you have at least onePICKAXE_STUDIO_*environment variable set in your MCP config.

Check that the studio name matches exactly (case-insensitive). Runstudios_listto see available options.

Your API key is invalid or expired. Get a new one from Pickaxe Studio settings.

Your API key doesn't have permission for this operation. Check your Pickaxe account permissions.

Contributions are welcome! Please seeCONTRIBUTING.mdfor guidelines.

- Pickaxe Platform
- Pickaxe Studio
- Model Context Protocol
- MCP Specification

Browse, search, and manage AI configuration blueprints and prompt hierarchies via LynxPrompt