RedM / RDR3 docs MCP server (`io.github.Cmoen11/redm-mcp`)

Model Context Protocol (MCP) server for RedM / RDR3 documentation. Provides native lookups (hash ↔ name), semantic search, framework docs for VORP, RSGCore, and oxmysql, plus access to rdr3_discoveries community data including peds, weapons, animations, AI flags, props, and audio banks. Runs as a hosted HTTP-transport MCP.

🛠️ Key Features

Native lookups (hash ↔ name)
Semantic search
Framework documentation: VORP, RSGCore, oxmysql
rdr3_discoveries community data (peds, weapons, animations, AI flags, props, audio banks)
Tool support: 10 tools

🚀 Use Cases

Retrieve exact RedM/RDR3 documentation answers for development workflows
Use community discovery data (assets and behavior flags) for reference during coding
Search across RedM/RDR3 docs and framework resources

⚡ Developer Benefits

HTTP-transport MCP endpoint at https://redm-mcp.fivem.no/mcp
Intended for AI coding agents (e.g., Claude Code, Cursor, Claude Desktop, VS Code Copilot, Zed)

⚠️ Limitations

Hosted service: no local install mentioned
No authentication indicated in the provided excerpt

redm-mcp — RedM / RDR3 docs MCP server

Model Context Protocol (MCP) server for RedM / RDR3: native lookups (hash ↔ name), semantic search, framework docs (VORP, RSGCore, oxmysql), and the rdr3_discoveries community data (peds, weapons, animations, AI flags, props, audio banks). Gives AI coding agents (Claude Code, Cursor, Claude Desktop, VS Code Copilot, Zed, …) exact answers instead of guesses.

Runs as a hosted HTTP-transport MCP at https://redm-mcp.fivem.no/mcp — no local install, no auth, just point your client at it.

This repository contains only installation/usage docs and serves as the issue tracker. The server source code is proprietary and not distributed. The hosted endpoint is free to use.

Why

When an agent reads RedM code it typically encounters:

lua

Citizen.InvokeNative(0x09C28F828EE674FA, player, 1.5, 5000)

Without a lookup, the agent guesses at parameters and semantics. With this server:

code

lookup_native({ hash: "0x09C28F828EE674FA" })
→ BOOST_PLAYER_HORSE_SPEED_FOR_TIME(player Player, speedBoost float, duration int)

The server advertises its own usage via MCP instructions — no skill or extra client config needed.

Tools

Tool	Use
`lookup_native`	Exact lookup by hash or name. O(1). Use for `Citizen.InvokeNative(0x...)` or `SCREAMING_SNAKE_CASE` names.
`semantic_search`	Search by behavior/concept ("teleport player", "spawn horse").
`grep_docs`	Regex/literal grep across raw doc files. Required for the large `rdr3_discoveries` data tables (audio_banks, ingameanims, …) which are only preview-indexed in embeddings.
`list_namespaces`	List categories and namespaces.
`browse`	List document paths under a category/namespace.
`get_document`	Fetch full markdown for a doc path.

Client setup

All clients connect to https://redm-mcp.fivem.no/mcp. No authentication required.

VS Code and Cursor users: click an install badge above for one-click setup. For everything else, copy-paste the config below — also available as ready-made files in examples/.

Claude Code

bash

claude mcp add --transport http redm-mcp https://redm-mcp.fivem.no/mcp

Or edit ~/.claude.json manually:

json

{
  "mcpServers": {
    "redm-mcp": {
      "type": "http",
      "url": "https://redm-mcp.fivem.no/mcp"
    }
  }
}

Cursor

Edit ~/.cursor/mcp.json (or .cursor/mcp.json in a project root):

json

{
  "mcpServers": {
    "redm-mcp": {
      "url": "https://redm-mcp.fivem.no/mcp"
    }
  }
}

Restart Cursor. Check Settings → MCP for green status.

Claude Desktop

Claude Desktop doesn't support HTTP MCP directly — use mcp-remote as a proxy. Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

json

{
  "mcpServers": {
    "redm-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://redm-mcp.fivem.no/mcp"
      ]
    }
  }
}

Restart Claude Desktop.

VS Code (Copilot / Continue)

.vscode/mcp.json in a project root:

json

{
  "servers": {
    "redm-mcp": {
      "type": "http",
      "url": "https://redm-mcp.fivem.no/mcp"
    }
  }
}

Zed

~/.config/zed/settings.json:

json

{
  "context_servers": {
    "redm-mcp": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "mcp-remote",
          "https://redm-mcp.fivem.no/mcp"
        ]
      }
    }
  }
}

Other clients (generic)

HTTP transport clients: point to https://redm-mcp.fivem.no/mcp
stdio-only clients: use mcp-remote as a proxy

Verify it works

curl https://redm-mcp.fivem.no/health → {"ok":true}
curl https://redm-mcp.fivem.no/ingest-status → shows version and last ingested timestamp
In your client: ask the agent to look up 0x09C28F828EE674FA — should return BOOST_PLAYER_HORSE_SPEED_FOR_TIME

Endpoints

Path	Auth	Purpose
`POST /mcp`	no	MCP protocol
`GET /health`	no	Liveness
`GET /ingest-status`	no	`{ current, last, ingestedAt, upToDate }`
`GET /stats`	no	Usage stats (tool counters, durations, breakdowns). Metadata only — no queries or secrets stored.
`GET /dashboard`	no	HTML dashboard over `/stats`.

Example prompts

See examples/prompts.md for things to try once installed (native lookups, behavior searches, framework / inventory questions, debugging).

Issues / feedback

Bug, missing docs, incorrect native lookups, requests for new tools? Open an issue.

Doc sources

Indexed upstreams:

femga/rdr3_discoveries — community ped/weapon/anim/prop hashes, AI flags, enums

query	string	required	Natural language or token query
category	string	optional	Limit to one doc category
namespace	string	optional	Limit to a native namespace, e.g. PLAYER, ENTITY
limit	integer	optional	How many ranked snippets to return. Default 20 (Anthropic contextual-retrieval research: top-20 outperforms top-5/10 before reranking).
mode	string	optional	Retrieval mode. Default hybrid (recommended).
responseFormat	string	optional	`concise` (default): 400-char snippet per hit — cheap, browse-style. `detailed`: full chunk content — use when you need an answer in one round-trip and want to skip the `get_document` follow-up.
prior_attempt	object	optional	Populate ONLY when retrying after a previous semantic_search call returned no useful results. Skip on first attempts.

category	string	optional
namespace	string	optional

path	string	required	Doc path. Two valid shapes: (a) `<category>/<file>.md` for docs, e.g. `vorp/vorp_core_docs.md`; (b) `learning:<id>` for community findings, e.g. `learning:11`. Use the path returned by `browse`/`semantic_search`/`grep_docs` verbatim — do not invent `learnings/<slug>.md`.
heading	string	optional	Optional prose heading from the doc, e.g. `Add Item to User` or `Character Management > Inventory Management`. Case-insensitive, fuzzy match on the leaf (text after the final `>`). NOT for code symbols — `addItem`, `getPlayerPed` etc. won't match; use `grep_docs` for those.

path	string	required	Doc path exactly as returned by `grep_docs` / `browse` / `semantic_search`, e.g. `discoveries/audio/audio_banks/audio_banks.lua`. Do not invent paths.
start	integer	required	First line to return (1-based, inclusive). Use the line number from a `grep_docs` hit.
end	integer	optional	Last line to return (1-based, inclusive). Omit for a 50-line window from `start`. Spans over 400 lines are capped.

pattern	string	required	Rust regex pattern (ripgrep engine). Case-insensitive by default. Prefer narrow, single-token patterns over kitchen-sink alternations.
caseInsensitive	boolean	optional	Default true. Set false for case-sensitive match.
category	string	optional	Limit to a doc category (e.g. discoveries, natives).
pathSubstring	string	optional	Substring filter on relative doc path, e.g. 'weapons' or 'clothes/cloth_hash_names'.
limit	integer	optional
contextBefore	integer	optional	Include N lines before each match (rg -B). Saves follow-up get_document calls when you need surrounding context.
contextAfter	integer	optional	Include N lines after each match (rg -A).
filesOnly	boolean	optional	Return only the list of matching paths (no per-line matches). Cheap for exploration before zoom-in.
multiline	boolean	optional	Allow `.` to match newlines and patterns to span lines (rg -U --multiline-dotall). Use for `(?s)foo.*bar` style.
prior_attempt	object	optional	Populate ONLY when retrying after a previous grep_docs call returned no matches. Skip on first attempts.

redm-mcp

RedM / RDR3 docs MCP server (`io.github.Cmoen11/redm-mcp`)

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

semantic_search

list_namespaces

browse

get_document

read_lines

grep_docs

lookup_native

asset_lookup

share_finding

get_invoke_guide

redm-mcp — RedM / RDR3 docs MCP server

Why

Tools

Client setup

Claude Code

Cursor

Claude Desktop

VS Code (Copilot / Continue)

Zed

Other clients (generic)

Verify it works

Endpoints

Example prompts

Issues / feedback

Doc sources

Install

Remote endpoint

Related servers

hash	string	optional	Native hash, e.g. 0x09C28F828EE674FA (case-insensitive, 0x optional)
name	string	optional	Native name, e.g. CAN_PLAYER_START_MISSION. Substring match if no exact hit.
namespace	string	optional	Restrict to a namespace, e.g. PLAYER, ENTITY. Only used with `name`.
limit	integer	optional

title	string	required	Short, searchable summary of the finding.
body	string	required	Markdown explaining WHY: context, the trap, the fix, verified behavior.
tags	array	optional	Up to 8 lowercase tags, e.g. ['weapons', 'damage'].
category	string	optional	Optional doc category this relates to.
source	string	optional	Optional short identifier of the contributing agent.

name	string	optional	Exact asset name, case-insensitive. Examples: `a_c_bear_01`, `weapon_pistol_volcanic`, `p_safe01`, `armysupplywagon`. Use when you know the precise name.
hash	string	optional	Asset hash (32-bit jenkins) in HEX format, case-insensitive, `0x` prefix optional. Examples: `0xBCFD0E7F`, `bcfd0e7f`. Use when you have a hash from decompiled code or another table and need the canonical name + metadata. Decimal-formatted hashes (e.g. `1946191463`) are NOT accepted — convert to hex first (`(1946191463).toString(16)`).
search	string	optional	Substring fragment within asset name, case-insensitive. Examples: `lemat`, `norfolk`, `volcanic`. Use when you remember part of the name. Algorithm: exact substring (ILIKE) first; if zero hits, falls back to pg_trgm `strict_word_similarity` ≥0.4 — catches abbreviation gaps like `moonshine`↔`_mshine_` when narrowed by `type` (without `type`, common terms find substring matches first and fuzzy never fires). `matchType` in the response tells you which path hit: `search` = exact substring, `fuzzy` = trigram.
type	string	optional	Filter results to one category. Useful when a name fragment matches multiple types (e.g. `horse` hits peds + vehicles).
limit	integer	optional	Max matches to return. Default 5, max 50. Only applies to `search` — exact `name`/`hash` always return 0 or 1.

RedM / RDR3 docs MCP server (io.github.Cmoen11/redm-mcp)

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

semantic_search

list_namespaces

browse

get_document

read_lines

grep_docs

lookup_native

asset_lookup

share_finding

get_invoke_guide

redm-mcp — RedM / RDR3 docs MCP server

Why

Tools

Client setup

Claude Code

Cursor

Claude Desktop

VS Code (Copilot / Continue)

Zed

Other clients (generic)

Verify it works

Endpoints

Example prompts

Issues / feedback

Doc sources

Install

Remote endpoint

Related servers

RedM / RDR3 docs MCP server (`io.github.Cmoen11/redm-mcp`)