Advanced Features

MCP Integration

Model Context Protocol (MCP) lets HexClaw agents call external tools and services.

Register an MCP Server

1. Open the Integration → MCP tab
2. Click "Add Server"
3. Provide the server name and connection endpoint
4. Select a transport: stdio, sse, or streamable_http
5. Connect and let HexClaw auto-discover available tools

Once connected, MCP tools can be assigned to agents. HexClaw shows live connection state as connected, disconnected, or error.

MCP Server Configuration Example

Below is a JSON configuration example for an MCP server:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@anthropic-ai/mcp-filesystem"],
      "env": { "ALLOWED_PATHS": "/Users/me/projects" }
    }
  }
}

Knowledge Base (RAG)

Upload documents to build a private knowledge base that the AI can retrieve from automatically during responses.

Supported Inputs

• PDF, Markdown, and TXT files
• Automatic chunking and vector embedding
• Semantic search retrieval

Workflow

1. Open the Knowledge Base page
2. Upload one or more documents
3. Wait for processing to move from processing to ready
4. In chat, HexClaw retrieves relevant chunks automatically as context

Semantic Memory

HexClaw can extract and store memory from conversations so useful context persists across sessions.

• Automatic capture - key details from conversations can be saved as memories
• Semantic search - in the Knowledge → Memory tab, search by keyword or semantic similarity
• Type filters - narrow results by memory type
• Manual management - review, delete, or bulk clean entries

Scheduled Jobs (Cron)

Schedule agents to run automatically at fixed times.

Create a Scheduled Job

1. Open the Automation → Scheduled Jobs tab
2. Click "Create Job"
3. Select an agent, enter a cron expression, and provide the prompt
4. Enable the job

# Cron expression examples
0 9 * * *      Every day at 9:00
0 */2 * * *    Every 2 hours
0 9 * * 1-5    Weekdays at 9:00

HexClaw also supports manual triggering, execution history, and heartbeat monitoring.

Skill Marketplace

Browse and install skills that extend what agents can do:

• Categories: communication, productivity, development, and data
• Installation: one-click install with per-agent enable or disable
• Ranking: sort by community rating and downloads
• Configuration: some skills expose custom parameters

Security Gateway

Configure these controls under Settings → Security:

Deep Thinking Mode

Switch to the Deep Thinking tab on the chat page, enter a question, and the agent runs multi-source deep analysis with a visible reasoning process:

1. Retrieve — gather relevant materials and sources
2. Analyze — perform in-depth analysis of each source
3. Synthesize — cross-reference and distill key insights, with citations

The reasoning process is shown in real time and can be collapsed or expanded to view the current stage.

Connections · IM Channels

HexClaw can connect to multiple instant messaging platforms, allowing you to chat with AI remotely via mobile messages. Go to the Connections hub (the "Connect" sidebar group) → Channels & Accounts to configure:

After saving the configuration, you must restart the Engine for changes to take effect. Each platform card has a "Test Connection" button to verify the configuration.

Data Connectors (GitHub / Notion)

Under Connections → Data Connectors, connect GitHub / Notion with a read-only token:

• Paste a PAT / integration token and click Test Connection to verify
• Browse resources directly (real repository / page listings)
• Tokens are encrypted at rest server-side (enc:v1:); API responses are always redacted

Webhook Notifications

Configure webhooks under Settings → Notifications to automatically push alerts when tasks complete, agents finish executing, or errors occur:

• WeCom — robot Webhook URL
• Feishu — custom robot Webhook
• DingTalk — group robot Webhook
• Custom — any HTTP POST endpoint

Each webhook can be configured with trigger events: task completion, agent completion, and error notifications.

ClawHub Skill Marketplace

In the Integration → Skills tab, browse and install skills from the OpenClaw community:

• Filter by category: programming, research, writing, data, automation, productivity
• Search by keyword for quick discovery
• One-click install; view installed skills in the "Installed" tab
• Each skill shows author, version, download count, and tags

You can also import skill files in .md format via local installation.

✏️ Edit this page on GitHub