Ceph Command Knowledge Base

Verified knowledge base of 1,254 Ceph CLI commands and 2,660 config parameters for Ceph 20.2.1 (Tentacle). Ships pre-generated — no Ceph cluster needed to use.

Exposes an MCP server and REST API so AI agents can verify commands, configs, and review test scripts before generating automation. Eliminates command hallucinations.

Setup

1. Clone and install

git clone https://github.com/pdhiran/ceph-command-kb.git
cd ceph-command-kb
pip install -e .

2. Connect your agent

Choose the integration that matches your agent:

Cursor — add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "ceph-cmd-kb": {
      "command": "python3",
      "args": ["-m", "ceph_command_kb.server.mcp_server", "--kb-path", "/path/to/ceph-command-kb/knowledge/ceph-20.2.1-tentacle"],
      "cwd": "/path/to/ceph-command-kb"
    }
  }
}

Restart Cursor. The MCP server starts automatically.

Claude Desktop / Continue / Cline / Windsurf / IBM Bob — all connect via SSE:

python -m ceph_command_kb.server.mcp_server --transport sse --host 0.0.0.0 --port 8081

Configure your agent to connect to http://localhost:8081/sse:

{
  "mcpServers": {
    "ceph-cmd-kb": {
      "url": "http://localhost:8081/sse",
      "transport": "sse"
    }
  }
}

For Claude Desktop: add to claude_desktop_config.json. For Bob: add to .bob/mcp.json. For Continue/Cline/Windsurf: use the MCP settings UI.

LangChain / CrewAI / CI pipelines / custom scripts — use the REST API:

python -m ceph_command_kb.server.rest_api --host 0.0.0.0 --port 9090

# Verify a command
curl -X POST http://localhost:9090/api/verify_command \
  -H "Content-Type: application/json" \
  -d '{"command": "ceph osd pool create"}'

# Verify a config parameter
curl -X POST http://localhost:9090/api/verify_config \
  -H "Content-Type: application/json" \
  -d '{"name": "osd_pool_default_size"}'

# Search commands
curl -X POST http://localhost:9090/api/search_commands \
  -H "Content-Type: application/json" \
  -d '{"query": "nfs cluster create"}'

# Review a test script
curl -X POST http://localhost:9090/api/review_test \
  -H "Content-Type: application/json" \
  -d '{"script_content": "ceph osd pool create mypool 32\nrbd create img --size 1024"}'

# Health check
curl http://localhost:9090/health

Additional integration guides:

• VS Code Extension — Inline command verification, search, and script review for VS Code / Bob users
• BOB_INTEGRATION_GUIDE.md — REST API integration guide with deployment options
• examples/bob_agent_integration.py — Ready-to-use Python client and LangChain/CrewAI tools

3. Use it

Once connected, agents automatically verify Ceph commands against the KB. You can also ask directly:

• "Verify the command `ceph osd pool create --size 3`"
• "What rbd commands are available for mirroring?"
• "Show me the default value of `osd_pool_default_size`"
• "Review this test file for issues"

Available Tools

Command Verification (12 tools)

Config Verification (4 tools)

Test Validation (2 tools)

Supported Ceph Binaries

ceph, rbd, rados, cephadm, ceph-volume, ceph-authtool, ceph-bluestore-tool, ceph-objectstore-tool, crushtool, monmaptool, osdmaptool

Running All Ceph MCPs Together

Three specialized MCPs work together as the Ceph Engineering Intelligence Platform:

Start all three for SSE clients (Bob, Claude Desktop, etc.):

python -m ceph_command_kb.server.mcp_server --transport sse --port 8081 &
python -m ceph_doc_kb.server.mcp_server --transport sse --port 8082 &
python -m ceph_issue_kb.server.mcp_server --transport sse --port 8083 &

Combined agent config (.bob/mcp.json, claude_desktop_config.json, etc.):

{
  "mcpServers": {
    "ceph-cmd-kb": { "url": "http://localhost:8081/sse", "transport": "sse" },
    "ceph-doc-kb": { "url": "http://localhost:8082/sse", "transport": "sse" },
    "ceph-issue-kb": { "url": "http://localhost:8083/sse", "transport": "sse" }
  }
}

Agents call whichever MCP has the right tools — the LLM decides automatically.

Further Documentation

• Development Guide — architecture, project structure, adding binaries, design decisions, maintainer guide