{
  "markdown": "<div align=\"center\">\n\n# Orion Agent Kit\n\n**Celestia network intelligence for AI agents.** An MCP server over [Orion](https://orionlabsone.cc/celestia)'s daily analysis: anomaly signals, network verdicts, staking &amp; DA metrics, live namespace and validator drill-downs — each figure linked back to the source it came from.\n\n[![MCP](https://img.shields.io/badge/MCP-server-6E56CF)](https://modelcontextprotocol.io)\n[![Celestia](https://img.shields.io/badge/Celestia-mainnet-7B2BF9)](https://celestia.org)\n[![Node](https://img.shields.io/badge/Node-%E2%89%A520-339933?logo=nodedotjs&logoColor=white)](https://nodejs.org)\n[![License: Apache-2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](./LICENSE)\n\n`https://orionlabsone.cc/mcp`\n\n</div>\n\n---\n\n## What this is\n\nOrion watches Celestia mainnet continuously and publishes a daily machine-written brief: what moved, what drifted from baseline, and why it matters. This kit exposes that analysis to AI agents over the [Model Context Protocol](https://modelcontextprotocol.io).\n\nThink of it as an **analysis layer** for the Celestia agent ecosystem. It sits alongside knowledge bases (which answer \"what is CIP-21?\") and raw-data or transaction kits (which list blobs or build a delegate tx). Orion answers a different question: **what is happening on the network right now, is it normal, and how did we get here?**\n\nEvery number the server returns carries a `provenance` link to the Celenium or Celestia LCD endpoint it was read from, so the agent can cite its sources — and so can you. The server is **read-only**: no wallets, keys, or transactions anywhere in it.\n\n## Tools\n\n**Today &amp; anomalies**\n\n| Tool | Question it answers |\n|---|---|\n| `get_daily_brief` | What's happening on Celestia today? (set `include_narrative` for Orion's written prose) |\n| `get_signals` | Any anomalies (info / notable / critical) today or on date X? |\n| `get_network_state` | Current height, DA volume, validators, staking ratios, Nakamoto coefficient, TIA price |\n\n**Trends &amp; history**\n\n| Tool | Question it answers |\n|---|---|\n| `get_trends` | What changed since yesterday / last week? (day- and week-over-week deltas) |\n| `get_metric_history` | Chart one metric (fees, blobs, price, staking ratio…) over N days |\n| `list_reports` | What daily reports exist in the archive? |\n| `get_report` | Full report for a given date |\n| `search_reports` | When did the archive last mention \"eclipse\" or \"block time\"? |\n\n**Live drill-down** (straight from Celenium)\n\n| Tool | Question it answers |\n|---|---|\n| `get_top_namespaces` | Which rollups post the most data to Celestia right now? |\n| `get_namespace` | Detail for one namespace: size, blob count, last activity |\n| `get_top_validators` | Largest validators by stake — moniker, commission, jailed status |\n\n## Resources &amp; prompts\n\nBeyond tools, the server exposes MCP **resources** an agent can read or subscribe to:\n\n- `orion://latest` — the most recent daily report\n- `orion://reports` — the archive index\n- `orion://report/{date}` — any archived report by date\n\n…and ready-made **prompts** to guide an agent:\n\n- `daily_celestia_briefing` — produce a grounded, source-cited daily briefing\n- `rollup_da_check` — survey who dominates blobspace and the current DA cost context\n\n## What counts as a signal\n\nOrion's detectors are **event-driven, not state-driven** — a persistently dominant namespace or a steady utilization day produces *no* signal, so what you see is what actually changed:\n\n- **DA utilization &amp; fees** — per-block volume and fees over the last day vs the recent daily baseline\n- **Blob-volume z-score** — statistical spikes/drops in data posted\n- **Namespace shift** — only when the DA leader changes or its share moves materially\n- **Staking &amp; concentration** — bonded-ratio moves, Nakamoto-coefficient changes, new jailings\n- **Market &amp; protocol** — notable TIA price moves, active governance proposals, new celestia-app releases\n\nEach daily brief is written by an LLM narrator and passes an automated **citation check** — every number in the prose is verified against the structured data before publishing.\n\n## Quick start — hosted endpoint (no install)\n\nThe MCP server runs at **`https://orionlabsone.cc/mcp`**.\n\n**Claude Code**\n\n```bash\nclaude mcp add --transport http orion https://orionlabsone.cc/mcp\n```\n\n**Cursor** — add to `.cursor/mcp.json` (already included if you cloned this repo):\n\n```json\n{ \"mcpServers\": { \"orion\": { \"url\": \"https://orionlabsone.cc/mcp\" } } }\n```\n\n**VS Code** — add to `.vscode/mcp.json`:\n\n```json\n{ \"servers\": { \"orion\": { \"type\": \"http\", \"url\": \"https://orionlabsone.cc/mcp\" } } }\n```\n\n**Claude Desktop / Codex / Windsurf / Cline / any stdio-only client** — bridge with [`mcp-remote`](https://www.npmjs.com/package/mcp-remote):\n\n```json\n{\n  \"mcpServers\": {\n    \"orion\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"mcp-remote\", \"https://orionlabsone.cc/mcp\"]\n    }\n  }\n}\n```\n\nPer-client walkthroughs: [Claude Code](docs/claude-code.md) · [Claude Desktop](docs/claude-desktop.md) · [Cursor](docs/cursor.md) · [Codex](docs/codex.md) · [VS Code](docs/vscode.md)\n\n## Quick start — clone &amp; chat\n\n```bash\ngit clone https://github.com/orioninfra/orion-agent-kit\ncd orion-agent-kit\nnpm install\nclaude   # Claude Code picks up .mcp.json automatically\n```\n\nThen just ask:\n\n> \"Any anomalies on Celestia today?\" · \"What changed on Celestia since yesterday?\" · \"Which rollups post the most data right now?\" · \"How concentrated is the validator set?\" · \"When did the archive last flag block time?\"\n\nThe bundled `/brief` and `/signals` commands wrap the two most common queries.\n\n## Run the server yourself\n\n```bash\nnpm install\nnpm run start          # Streamable HTTP on 127.0.0.1:8791/mcp\nnpm run start:stdio    # stdio transport (for local MCP clients)\nnpm run smoke          # end-to-end check of the tools against the live API\nnpm run typecheck      # type-check the sources\n```\n\nEnvironment: `PORT` (default `8791`), `HOST` (default `127.0.0.1`), `ORION_API_BASE` (default `https://orionlabsone.cc`), `CELENIUM_BASE` (default `https://api-mainnet.celenium.io/v1`, used by the drill-down tools).\n\n## REST, without MCP\n\nThe underlying API is plain JSON, usable directly or via function calling with the [OpenAPI spec](https://orionlabsone.cc/openapi.json):\n\n| Endpoint | Returns |\n|---|---|\n| `GET /api/latest` | today's full report |\n| `GET /api/reports` | archive index |\n| `GET /api/reports/{date}` | full report for a date |\n| `GET /api/rss` | the brief as an RSS feed |\n\n## How it fits together\n\n```\nAI agent (Claude / GPT / any MCP client)\n   │  MCP — read-only analysis tools, resources & prompts\n   ▼\norion-agent-kit  ──►  Orion daily analysis (orionlabsone.cc/api)\n        │                 │  every figure linked to its source\n        │                 ▼\n        │             Celestia LCD · market data\n        └──────────►  Celenium API   (live namespace & validator drill-down)\n```\n\n## License\n\n[Apache-2.0](./LICENSE), Copyright 2026 Orion Labs. Questions, ideas, or a tool you wish existed: open an issue.\n",
  "bytes": 7085,
  "sha": "96b1114cc74ea8327d880bb43c49a5df73e56b78aa171691754214d032bc0d70",
  "repo_slug": "orioninfra/orion-agent-kit",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_cc_orionlabsone_orion_agent_kit_3a002769/readme"
}