Meteora 백서

Meteora

원문

bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 75쪽 중 1쪽

# llms.txt Source: https://docs.meteora.ag/agents/llms-txt Static documentation indexes and raw Markdown endpoints for LLMs and AI agents. Meteora provides structured documentation files purpose-built for LLMs and AI agents, following the [llmstxt.org](https://llmstxt.org) standard. <Note> Use `llms.txt` when your agent needs a compact map of the docs. Use [Documentation MCP](/agents/mcp) when your AI tool can call MCP tools and should search or read pages interactively. </Note> ## llms.txt A plain Markdown file containing a structured index of Meteora documentation with titles and descriptions for each indexed page: * **Site title** as an H1 heading * **Sections** for major areas such as Get Started, User Guides, Core Products, Helper Products, Developer Guides, Invent, Agents, Protocol, and Resources * **Links with descriptions** for product guides, SDK references, API references, and resources * **OpenAPI spec links** where API reference specifications are indexed <Card title="llms.txt" href="https://docs.meteora.ag/llms.txt" icon="file"> Open the llms.txt for Meteora docs </Card> ```mdx theme={"system"} # Meteora Documentation ## Get Started - [We Build Liquidity Pools](https://docs.meteora.ag/get-started.md): The Most Composable Liquidity Layer... - [What is DLMM?](https://docs.meteora.ag/core-products/dlmm/what-is-dlmm.md): Learn how Meteora's DLMM organizes liquidity... - [DAMM v2 Developer Guide](https://docs.meteora.ag/developer-guides/damm-v2/index.md): Learn how to build DAMM v2 integrations... - [Dynamic Bonding Curve Developer Guide](https://docs.meteora.ag/developer-guides/dbc/index.md): Learn how to build Dynamic Bonding Curve integrations... ## OpenAPI Specs - [openapi](https://docs.meteora.ag/api-reference/dlmm/openapi.json) ``` This is the best first fetch for agents that need to decide which Meteora product, guide, SDK reference, API reference, or raw Markdown page to read next. ## llms-full.txt <Card title="llms-full.txt" href="https://docs.meteora.ag/llms-full.txt" icon="file"> Open the llms-full.txt for Meteora docs </Card> While `llms.txt` provides a concise index, `llms-full.txt` contains the **entire documentation site** as context, including descriptions, code examples, guides, API references, and product explanations. Use `llms-full.txt` when: * Your AI tool needs **complete, granular context** for deep indexing * You're building **custom RAG pipelines** over Meteora docs * You want **every code example** available for reference ## Raw Markdown Pages Any documentation page can be accessed as raw Markdown, making it easy for AI agents to consume individual pages programmatically. **Method 1: Append `.md` to the URL** ```bash theme={"system"} curl https://docs.meteora.ag/core-products/dlmm/what-is-dlmm.md curl https://docs.meteora.ag/core-products/damm-v2/what-is-damm-v2.md curl https://docs.meteora.ag/core-products/dbc/what-is-dbc.md curl https://docs.meteora.ag/developer-guides/dlmm/typescript-sdk/getting-started.md curl https://docs.meteora.ag/developer-guides/damm-v2/typescript-sdk/getting-started.md curl https://docs.meteora.ag/developer-guides/dbc/typescript-sdk/getting-started.md ``` **Method 2: Use the `Accept` header** ```bash theme={"system"} curl -H "Accept: text/markdown" https://docs.meteora.ag/developer-guides/damm-v2/typescript-sdk/getting-started ``` Both methods return the page content as `text/markdown`. ## How Agents Should Use These Files | Need | Best entrypoint | | :- | :- | | Discover the docs structure | `https://docs.meteora.ag/llms.txt` | | Ingest the full docs corpus | `https://docs.meteora.ag/llms-full.txt` | | Read one exact page | Append `.md` to the page URL | | Search and read from an MCP client | `https://docs.meteora.ag/mcp` | <Tip> For most agent workflows, start with `llms.txt`, fetch only the relevant `.md` pages, and use `llms-full.txt` only when you intentionally want the full corpus in one document. </Tip> # Documentation MCP Source: https://docs.meteora.ag/agents/mcp Connect AI tools to Meteora docs through Mintlify's hosted Model Context Protocol server. Meteora's documentation MCP is the live, tool-call version of these docs for AI agents. It is hosted by Mintlify at `https://docs.meteora.ag/mcp` and follows Mintlify's documentation MCP model: agents can search indexed documentation, then read exact page content from a read-only docs filesystem. ## What is MCP? MCP is an open protocol that connects AI assistants to external data sources and tools. With Meteora's documentation MCP server, your AI tool can: * **Search documentation**: find relevant product docs, developer guides, SDK references, and API references * **Read exact pages**: retrieve full MDX/Markdown content through Mintlify's virtual docs filesystem * **Inspect API reference content**: look up indexed endpoint docs, request/response schemas, parameters, and errors where available * **Use docs without web-search noise**: answer from the current indexed Meteora documentation instead of generic search results <Note> This is a read-only documentation server. It does not execute API calls, submit transactions, sign with a wallet, mutate documentation, or manage funds. </Note> ## Endpoint Meteora's public documentation MCP server is available at: ``` https://docs.meteora.ag/mcp ``` <Card title="Meteora Docs MCP" href="https://docs.meteora.ag/mcp" icon="plug"> Connect an MCP-compatible AI tool to Meteora documentation </Card> This endpoint uses HTTP transport and exposes Meteora's indexed documentation through MCP. ## Documentation MCP vs Mintlify MCP Mintlify offers two related MCP experiences: | Server | Audience | What it can do | | :- | :- | :- | | **Documentation MCP** at `https://docs.meteora.ag/mcp` | Meteora builders and users | Search and read Meteora's published docs | | **Mintlify MCP** at `https://mcp.mintlify.com` | Trusted documentation teams | Edit Mintlify content, update navigation, modify `docs.json`, and open pull requests | Use Meteora's documentation MCP when an agent needs current Meteora context. Use Mintlify's authenticated MCP only if you are part of the docs team and want an agent to make documentation changes through Mintlify's branch and pull request workflow. ## What You Get Once connected, your AI assistant can search and read: | Resource | Description | | :- | :- | | **Documentation pages** | Product explanations, user guides, developer guides, and reference docs | | **API references** | Indexed endpoint docs and OpenAPI-backed request/response details where available | | **Program references** | Account, instruction, event, and error documentation for Meteora programs | | **Skill resources** | `skill.md` resources, when available through the docs site | <Tip> Combine MCP with [llms.txt](/agents/llms-txt): use MCP for in-editor queries and use `llms.txt`, `llms-full.txt`, or raw Markdown exports for batch processing and RAG pipelines. </Tip> ## Setup <Tabs> <Tab title="Claude Code"> **How to enable Meteora Docs MCP in Claude Code:** 1. Open your terminal. 2. Add Meteora's MCP server by running: ```bash theme={"system"} claude mcp add --transport http meteora https://docs.meteora.ag/mcp ``` 3. Launch Claude Code and start chatting with Meteora. **Or add to your project’s `.mcp.json`:** ```json theme={"system"} { "mcpServers": { "meteora": { "url": "https://docs.meteora.ag/mcp" } } } ``` **Verify with:** ```bash theme={"system"} claude mcp list ``` </Tab> <Tab title="Claude"> **Setup Meteora Docs MCP in Claude:** 1. Go to [**Connectors**](https://claude.ai/settings/connectors) in your Claude settings. 2. Select **Add custom connector**. 3. Set the name to **Meteora** and the URL to **[https://docs.meteora.ag/mcp](https://docs.meteora.ag/mcp)**. 4. When chatting, click the attachments button (plus icon) and select the Meteora connector. </Tab> <Tab title="Cursor"> **Setup Meteora Docs MCP in Cursor:** 1. Open the command palette with `Cmd + Shift + P` (or `Ctrl + Shift + P` on Windows). 2. Search for **Open MCP settings**. 3. Add the following to your `mcp.json` file: ```json theme={"system"} { "mcpServers": { "meteora": { "url": "https://docs.meteora.ag/mcp" } } } ``` 4. Save and reload Cursor. </Tab> <Tab title="VS Code"> **Setup Meteora Docs MCP in VS Code:** 1. Create or open `.vscode/mcp.json`. 2. Add the following server configuration: ```json theme={"system"} { "servers": { "meteora": { "type": "http", "url": "https://docs.meteora.ag/mcp" } } } ``` 3. Save the file and reload your MCP-enabled VS Code environment. </Tab> <Tab title="Windsurf"> **Using Meteora Docs MCP in Windsurf:** 1. Open the command palette with `Cmd + Shift + P` (or `Ctrl + Shift + P` on Windows). 2. Search for **Windsurf: Configure MCP Servers**. 3. Add the following to your `mcp_config.json` file: ```json theme={"system"} { "mcpServers": { "meteora": { "serverUrl": "https://docs.meteora.ag/mcp" } } } ``` 4. Save and restart Windsurf. </Tab> </Tabs> ### Other MCP-Compatible Tools Any tool that supports HTTP MCP servers can connect to `https://docs.meteora.ag/mcp`. ## Example Prompts ```text theme={"system"} Use the Meteora Docs MCP to find the current DLMM TypeScript SDK swap flow, then summarize the files I should read before implementing it. ``` ```text theme={"system"} Search Meteora docs for DAMM v2 fee scheduler and rate limiter behavior. Cite the relevant docs pages before suggesting configuration values. ``` ```text theme={"system"} Read the DBC migration docs and program error reference, then help me design a defensive integration checklist. ``` <Note> For onchain execution, keep signing and transaction submission in your own scripts, wallet, SDK integration, or [Meteora Invent](/invent/actions). The documentation MCP supplies context; it does not perform the action. </Note> # Agents Source: https://docs.meteora.ag/agents/overview Use AI agents with Meteora docs, APIs, SDKs, and launch tooling. <video /> Meteora is built to be legible to AI agents: docs are available as MCP tools, static LLM indexes, raw Markdown, SDK guides, API references, and command-driven launch workflows. Use the docs layer for context, then execute through your own code, wallet, SDK integration, or Meteora Invent when an action is intentional. ## Agent Workflow 1. **Discover** the relevant product, guide, SDK, or API page with [llms.txt](/agents/llms-txt) or [Documentation MCP](/agents/mcp). 2. **Read exact context** from raw Markdown pages or MCP page retrieval before writing code. 3. **Implement** against Meteora SDKs, REST API references, and program docs. 4. **Execute deliberately** through scripts, wallets, SDK transactions, or [Meteora Invent](/invent/actions), with signing controlled by the user. For agents that support skills, install the [Meteora Agent Skill](/agents/skill) — it packages the full action playbook with built-in safety gates. <Note> Meteora's documentation MCP is read-only. It helps agents understand Meteora; it does not sign transactions, manage private keys, execute swaps, create pools, or edit this documentation. </Note> ## Pick the Right Entry Point <CardGroup> <Card title="Ask questions in an AI editor or hosted chat" icon="comments" href="/agents/mcp"> Use Documentation MCP when the agent supports MCP and should search or read current docs in context. </Card> <Card title="Build a docs index or RAG pipeline" icon="database" href="/agents/llms-txt"> Use `llms.txt` when you need static, crawlable documentation context. </Card> <Card title="Fetch one page exactly" icon="file-lines"> Use a raw Markdown export when you already know the page URL and want clean Markdown. </Card> <Card title="Write integration code" icon="code" href="/developer-guides/dlmm"> Use Developer Guides when you need SDK setup, account models, instructions, events, errors, and examples. </Card> <Card title="Use indexed REST data" icon="server" href="/developer-guides/dlmm/api-reference/overview"> Use API Reference pages when you need endpoint schemas, parameters, response fields, and rate-limit context. </Card> <Card title="Launch or test onchain workflows" icon="terminal" href="/invent/actions"> Use Meteora Invent when you need config-driven commands for pools, launches, or protocol actions. </Card> <Card title="Give your agent the full playbook" icon="wand-magic-sparkles" href="/agents/skill"> Install the Meteora Agent Skill when your agent runtime supports Agent Skills and should execute Meteora actions end to end with safety gates. </Card> </CardGroup> ## Agent Surfaces For agents that need to discover, reason, implement, or prepare actions with Meteora: <CardGroup> <Card title="Documentation MCP" icon="plug" href="/agents/mcp"> Search and read Meteora docs from Claude, Claude Code, Cursor, VS Code, Windsurf, ChatGPT, and other MCP-compatible tools. </Card> <Card title="llms.txt" icon="file-lines" href="/agents/llms-txt"> Static LLM-optimized indexes and raw Markdown exports for agents, crawlers, and custom retrieval pipelines. </Card> <Card title="Developer Guides" icon="code" href="/developer-guides/dlmm"> SDKs, program references, API references, and examples for building Meteora integrations. </Card> <Card title="Meteora Invent" icon="terminal" href="/invent/actions"> Config-driven commands for launching pools and running supported onchain workflows with user-controlled signing. </Card> </CardGroup> ## Local and Hosted Agents <CardGroup> <Card title="Local coding agents" icon="laptop-code"> For Claude Code, Cursor, Codex, VS Code, and Windsurf, connect Documentation MCP, read raw Markdown when needed, install SDKs locally, run tests, and execute scripts only with explicit wallet and network configuration. </Card> <Card title="Hosted agents" icon="cloud"> For Claude.ai, ChatGPT, and other MCP clients, connect Documentation MCP for current docs context, use `llms.txt` links for fallback retrieval, and keep wallet signing or production execution outside the chat unless a trusted execution tool is explicitly configured. </Card> </CardGroup> ## Safe Agentic Usage * Ask the agent to cite the Meteora docs pages it used before implementing. * Never paste seed phrases, private keys, or production wallet secrets into an agent chat. * Use localnet or devnet before mainnet when testing generated scripts. * Keep transaction review and signing under user control. * Treat large automated changes or onchain actions as proposals until reviewed. ## Prompt Starters ```text theme={"system"} Use the Meteora Docs MCP to find the current DAMM v2 TypeScript SDK guide, then draft the smallest safe integration plan for adding liquidity. ``` ```text theme={"system"} Start from Meteora llms.txt, identify the DBC pages needed for migration handling, then fetch only those Markdown pages before writing code. ``` # Meteora Agent Skill Source: https://docs.meteora.ag/agents/skill Install and use Meteora's Agent Skill for onchain actions and SDK development. The Meteora Agent Skill is a portable set of instructions for AI coding agents. It helps an agent choose the right Meteora product, collect the required inputs, run supported studio commands, and write TypeScript against the SDK APIs the repository currently uses. The skill lives at [`skills/meteora`](https://github.com/MeteoraAg/meteora-invent/tree/main/skills/meteora) in the `meteora-invent` repository and follows the open [Agent Skills format](https://agentskills.io). You can install the same folder in Claude Code, OpenClaw, Hermes, Cursor, or another compatible runtime. <Note> An Agent Skill is an instruction pack, not a hosted transaction service or wallet. The agent still needs terminal and network access to run commands, and you remain in control of the wallet and signing environment. </Note> ## When to Use It The skill separates requests into two paths: | Path | Use it when | What the agent uses | | :- | :- | :- | | **ACT** | You want to launch, swap, create a pool, manage liquidity, claim fees, or run another supported action | The config-driven CLI in `meteora-invent/studio` | | **BUILD** | You want a bot, backend, frontend, custom transaction flow, or an operation that the studio does not expose | Meteora's TypeScript SDKs and the skill's version-specific reference packs | For current product concepts and API documentation, use the [Documentation MCP](/agents/mcp) or [llms.txt](/agents/llms-txt) alongside the skill. The skill provides the workflow and safety rules; the docs provide the broader protocol reference. ## What the Skill Covers ### Studio Actions The ACT path documents 81 commands from the Meteora Invent studio. Each entry records the command, accepted selector flags, config block, important units, expected output, and approximate SOL requirement. | Area | Commands | Common uses | | :- | :- | :- | | Setup | 2 | Convert a private key into the studio keypair file and request devnet or localnet SOL | | DBC | 8 | Create configs and pools, swap, claim trading fees, inspect graduation status, transfer the pool creator, and migrate to DAMM v1 or DAMM v2 | | DLMM | 10 | Create and seed pools, control pool status, place or cancel limit orders, swap, read positions, and claim fees and rewards | | DAMM v2 | 10 | Create balanced or one-sided pools, add or remove liquidity, split or close positions, refresh vesting, swap, and read positions | | DAMM v1 | 5 | Create legacy pools, lock LP, create Stake2Earn farms, lock LP into those farms, and swap | | Alpha Vault | 7 | Create a vault, deposit, withdraw, claim, withdraw remaining quote, crank the fill, and read vault and escrow state | | Presale Vault | 9 | Create a presale, deposit, withdraw, claim, sweep remaining quote, withdraw the raise, handle unsold supply, close an escrow, and read status | | Met Lock | 5 | Create a vesting escrow, attach escrow metadata, claim vested tokens, and read one or all escrows for a wallet | | Stake2Earn | 6 | Stake into a fee farm, claim fees, start or cancel an unstake, withdraw after the cooldown, and read farm and staker state | | Pool Farms | 5 | Stake DAMM v1 LP, unstake, claim rewards from one farm or several, and read farm and user state | | Dynamic Vault | 3 | Deposit into a lending vault, withdraw, and read supply, withdrawable amount, and virtual price | | Fee Sharing | 8 | Create a fee vault, fund it directly or from DAMM v2 fees, rewards, or DBC fees, transfer a position to the vault, claim a share, and read the split | | Zap | 3 | Enter a DAMM v2 or DLMM position with a single token and exit a position into one token | The studio uses one fixed JSONC config file per protocol. Command-line flags select an account such as a mint, pool, or limit order; they do not select a different config file. For example: ```bash Terminal theme={"system"} pnpm studio dbc-get-status --baseMint <MINT> pnpm studio dlmm-swap --poolAddress <POOL> pnpm studio damm-v2-get-positions --poolAddress <POOL> ``` Before a write, the agent edits the relevant file under `studio/config`, keeps `dryRun` set to `true`, and shows you the simulation result. It changes `dryRun` to `false` only after you confirm the real execution and the target network, then returns it to `true` after the action. ### SDK Development The BUILD path contains focused references for the four main liquidity protocols: | Product | Package and pinned version | Use it for | | :- | :- | :- | | DBC | `@meteora-ag/dynamic-bonding-curve-sdk@1.5.11` | Token launches, curve configuration, trading, fee claims, graduation, and migration | | DAMM v2 | `@meteora-ag/cp-amm-sdk@1.4.5` | Constant-product pools, position NFTs, swaps, liquidity, locks, vesting, fees, and rewards | | DLMM | `@meteora-ag/dlmm@1.9.14` | Bin-based liquidity, strategies, swaps, positions, fee claims, limit orders, and rebalanci
프로젝트로 돌아가기