About Hoodly GMGN Robinhood Research
Hoodly GMGN Robinhood Research is a read-only MCP server for researching tokens on Robinhood Chain through GMGN. It serves Hoodly's financial-agent stack with launch discovery, token analysis, and market context, without trading or wallet-signing tools.
Key features
- 01Discover new and near-completion launches
- 02Screen indexed launchpads for quality
- 03View Robinhood Chain token rankings
- 04Inspect token security, pools, holders, and traders
- 05Compare research data for 2–5 tokens
- 06Retrieve OHLCV and momentum context
Use cases
- 01Find launchpad tokens for further review
- 02Assess token security and holder distribution
- 03Compare several tokens before presenting findings
MCP support
Exposes read-only Robinhood Chain discovery and research tools to MCP clients, backed by the GMGN CLI/API. It has no swap, order, signing, wallet, or private-key tools.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://gmgn.hoodly.io/mcpclaude mcp add --transport http hoodly-agent-hoodly-gmgn-robinhood https://gmgn.hoodly.io/mcp{"mcpServers":{"hoodly-agent-hoodly-gmgn-robinhood":{"type":"http","url":"https://gmgn.hoodly.io/mcp"}}}Open in Cursor{"servers":{"hoodly-agent-hoodly-gmgn-robinhood":{"type":"http","url":"https://gmgn.hoodly.io/mcp"}}}Open in VS Code{"mcpServers":{"hoodly-agent-hoodly-gmgn-robinhood":{"type":"http","url":"https://gmgn.hoodly.io/mcp"}}}{"mcpServers":{"hoodly-agent-hoodly-gmgn-robinhood":{"serverUrl":"https://gmgn.hoodly.io/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
10| Parameter | Type | Required | Description |
|---|---|---|---|
| limit | integer | No | |
| stage | enum | No | new_creation | near_completion | completed |
| launchpads | array | No | Restrict to GMGN launchpad platforms on Robinhood; omit for all launchpads. |
| minLiquidityUsd | number | No |
Details
From the project READMEHoodly GMGN Robinhood Research MCP
Read-only MCP service for Hoodly's Robinhood financial-agent stack. It uses the GMGN CLI/API for chain-wide launchpad discovery and token due diligence. It deliberately contains no swap, signing, wallet, or private-key tool.
Tools
hoodly_list_robinhood_launches— newly created / near-completion / graduated launchpad tokens.hoodly_screen_robinhood_launchpads— server-side quality screen across every indexed launchpad.hoodly_robinhood_trending— Robinhood Chain market ranking.hoodly_research_robinhood_token— info, security, pool, holders, leading traders, plus deterministic research score.hoodly_robinhood_token_kline— OHLCV/momentum context.hoodly_compare_robinhood_tokens— comparable research packets for 2–5 addresses.hoodly_robinhood_daily_brief— daily briefing input for the Hoodly API scheduler.
The launchpad feed is Robinhood-wide. If the product needs a Pons-only view, add a verified Pons provenance/indexer source; do not label GMGN-only results as Pons launches.
Hoodly financial stack
Hoodly Web (Next.js)
→ Hoodly API (Express): agent permissions, sessions, saved briefs
→ this MCP (read-only research boundary)
→ GMGN CLI/API, chain=robinhood
Keep GMGN_API_KEY only in this service or the Hoodly API secret manager. Do not set GMGN_PRIVATE_KEY.
The matching local workflow is SKILL.md. GMGN portfolio can be added later only for a user-authorized, read-only wallet view. Do not expose GMGN swap/order/cooking endpoints; GMGN's KOL/smart-money tracking is not currently documented for Robinhood.
Example Hoodly agent policy
- Call
hoodly_list_robinhood_launchesorhoodly_robinhood_daily_brief. - Select a contract address, never a token symbol alone.
- Call
hoodly_research_robinhood_token. - Present observed metrics, derived score, missing evidence, and risks to the user.
- Keep the final action as user decision. No execution tool exists in this MCP.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.