{
  "markdown": "# @dcp/mcp — DCP Model Context Protocol server\n\n**DCP** is Saudi Arabia's sovereign AI compute platform — in-Kingdom inference, GPU rental, and persistent storage, billed in SAR, on Saudi-owned hardware. This MCP server lets any MCP-capable agent (Claude Desktop, Claude Code, Cursor, or your own) use DCP through native tool calls.\n\nIt is a thin client: it reads `DCP_API_KEY` from the environment and calls `https://api.dcp.sa`. No secrets are bundled.\n\n## Install (one line)\n\nThe connector runs straight from GitHub — no npm install, no clone:\n\n```bash\nDCP_API_KEY=dc1-sk-xxxxxxxx npx -y github:dhnpmp-tech/dcp-mcp\n```\n\n## MCP client config\n\nAdd this to your MCP client (Claude Desktop `claude_desktop_config.json`, Claude Code / Cursor `.mcp.json`):\n\n```json\n{\n  \"mcpServers\": {\n    \"dcp\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"github:dhnpmp-tech/dcp-mcp\"],\n      \"env\": { \"DCP_API_KEY\": \"dc1-sk-xxxxxxxx\" }\n    }\n  }\n}\n```\n\nThen ask your agent to \"list DCP models\" or \"rent a GPU on DCP for 30 minutes.\" Agents with no key can call `register_agent` first — see the quickstart below.\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `register_agent` | Self-register in one call — get a real API key + small SAR trial, **no human, no email click**. Use first when you have no key. |\n| `list_models` | List currently-serveable models (OpenAI-style). |\n| `chat` | Run a chat completion (OpenAI-compatible, in-Kingdom). |\n| `get_balance` | Renter wallet balance (SAR). |\n| `list_gpus` | List rentable GPU **types** (e.g. `H100`, `RTX 4090`) with VRAM + live availability. |\n| `create_pod` | Rent a whole GPU as an interactive pod (root + Jupyter + SSH), prepaid per minute. |\n| `get_pod` | Pod status + access URL / SSH command / time remaining. |\n| `extend_pod` | Add time to a running pod, no restart. |\n| `stop_pod` | Stop early; unused prepaid time refunded. |\n| `rent_volume` | Rent an exclusive in-Kingdom persistent volume (10/20/30 GB). |\n| `get_volume` | Active volume status + usage. |\n\n## Zero-human quickstart (for agents)\n\nThe whole loop runs with no human in it:\n\n1. **`register_agent`** → returns a real `api_key` + a **20 SAR** trial credit (no email click). Set it as `DCP_API_KEY`.\n2. **`list_gpus`** → pick a `gpu_type` string (e.g. `\"H100\"`, `\"RTX 4090\"`) from the live, available types.\n3. **`create_pod`** with that `gpu_type` + `duration_minutes` → poll **`get_pod`** for the `access_url` / `ssh_command` once running.\n4. **`chat`** → run OpenAI-compatible inference on an available model from `list_models`.\n5. **`stop_pod`** → stop early; unused prepaid minutes are refunded to the wallet.\n\nMinting a key by hand (equivalent to `register_agent`):\n\n```bash\ncurl -s -X POST https://api.dcp.sa/api/renters/agent-register \\\n  -H 'Content-Type: application/json' -d '{}'\n# → { \"api_key\": \"dcp-renter-…\", \"trial_credit_sar\": 20, \"balance_sar\": 20, ... }\n```\n\nThe trial (20 SAR) is enough to list GPUs, run a short pod, and do real inference; the larger grant stays behind email-verified signup. Calls are per-IP rate-limited.\n\n## Environment\n\n- `DCP_API_KEY` — renter API key (required for every tool except `register_agent`).\n- `DCP_API_BASE` — API host, default `https://api.dcp.sa`.\n\n## Why DCP\n\nInference, GPU rental, fine-tune hosting, and storage on Saudi-owned hardware inside the Kingdom — full PDPL / data-residency compliance, billed in SAR. The inference API is a drop-in OpenAI replacement: point any OpenAI SDK at `https://api.dcp.sa/v1`.\n\nLearn more: **https://dcp.sa/v2/agents** · `https://dcp.sa/llms.txt`\n\nMIT licensed.\n",
  "bytes": 3564,
  "sha": "64eeb88ddf197c9915b4b73cda4efe0fd87194626583de8cbe7d7b20266a1a0e",
  "repo_slug": "dhnpmp-tech/dcp-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_dhnpmp_tech_dcp_mcp_25fae1a1/readme"
}