Lifi logo

Lifi

Organization
sendaifun
lifi

Integrate LI.FI for cross-chain swaps, bridging, payments, route discovery, and transfer status tracking across Solana, EVM, Bitcoin, and Sui. Use when building Solana applications or AI agents that need quotes, routes, executable transactions, supported chains/tokens/tools, or cross-chain transfer monitoring.

Overview

Publishersendaifun
Repositoryskills
Skill namelifi
Stars
128
Forks
81
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by sendaifun on GitHub. Read the source before you install it.

Installation

Install the Lifi AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/sendaifun/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/lifi .claude/skills/lifi
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Lifi in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Lifi on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Lifi is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

LI.FI Cross-Chain Integration Guide

Use LI.FI when a Solana app, wallet, backend, or AI agent needs cross-chain token transfers, bridge aggregation, same-chain swaps, payment flows, or transfer status tracking. LI.FI exposes the same routing engine through REST API, SDK, MCP server, CLI, and Widget surfaces.

Overview

LI.FI is a multi-chain liquidity aggregation platform for swaps and bridging. For agent and backend integrations, the REST API is the lowest-dependency default. For frontend apps, prefer the SDK because it handles wallets, signing, execution tracking, and ecosystem-specific transaction handling. For MCP-compatible AI hosts, prefer the LI.FI MCP server for typed tool discovery.

Core capabilities:

  • Quotes: Get a ready-to-sign transaction for the best route.
  • Routes: Compare multiple route options and execute step-by-step.
  • Status tracking: Track source and destination chain transfer progress.
  • Discovery: Query supported chains, tokens, bridges, and exchanges at runtime.
  • Solana support: Request Solana-specific chain/token data via chainTypes=SVM and use SOL as the Solana chain key in API examples.

Product Surface Selection

Choose the smallest surface that matches the task:

  • REST API: Backend services, scripts, and general AI agents. Simple HTTP calls, no runtime dependency.
  • SDK (@lifi/sdk): Frontend or full execution flows. Handles wallet connectors, signing, route execution, and update hooks.
  • MCP Server: MCP-compatible AI hosts such as Claude, Cursor, or Windsurf. Use when typed tool discovery is available.
  • CLI: Token-efficient agent workflows where compact human-readable output is preferable to raw JSON.
  • Widget: Ready-made UI when the user wants embedded swap/bridge UX rather than custom integration.

Do not force every integration through the SDK. For quote lookup, route comparison, or status checks, the API is often simpler and easier to audit.

Base URLs and Authentication

text
Production API: https://li.quest/v1
Staging API:    https://staging.li.quest/v1
Docs:           https://docs.li.fi
OpenAPI:        https://docs.li.fi/openapi.yaml
LLM overview:   https://docs.li.fi/llms.txt

LI.FI APIs can be used without an API key. Use an API key for higher rate limits or authenticated partner usage. Register an integration in the LI.FI Partner Portal to get an API key: https://portal.li.fi/

For direct REST calls, pass the key in the x-lifi-api-key header:

bash
curl 'https://li.quest/v1/chains?chainTypes=EVM,SVM' \
  --header 'accept: application/json' \
  --header 'x-lifi-api-key: YOUR_API_KEY_IF_AVAILABLE'

Test a key server-side before using it in production:

bash
curl 'https://li.quest/v1/keys/test' \
  --header 'x-lifi-api-key: YOUR_API_KEY'

Never expose x-lifi-api-key in browser code, public repositories, or direct Widget configuration. If using the SDK from a backend or trusted runtime, pass the key through createConfig({ apiKey: '...' }); if using the Widget in a frontend, do not pass an API key.

Integration Workflow

  1. Clarify the transfer intent

    • Source chain, destination chain, source token, destination token.
    • Amount in the token's smallest unit.
    • Sender address and, when different, recipient address.
    • Whether the user wants a single best route or route comparison.
  2. Discover support instead of hardcoding

    • Use /chains to verify chains.
    • Use /tokens or /token to verify token addresses and decimals.
    • Use /tools to list current bridges and exchanges.
    • Do not assume every token, bridge, or chain pair is available.
  3. Choose quote vs routes

    • Use GET /quote for a simple transfer where the best executable route is enough.
    • Use POST /advanced/routes when comparing alternatives or when the user asks for route choice, cost, speed, tool allowlists, or multiple steps.
    • Use POST /advanced/stepTransaction to populate transaction data for individual route steps when executing advanced routes.
  4. Show the user what they will sign

    • Summarize from-chain, to-chain, from-token, to-token, amount, estimated output, tool/bridge, fees, recipient, and slippage.
    • Never ask a user to sign opaque transaction data without a human-readable summary.
  5. Execute through the appropriate wallet path

    • If using GET /quote, the response already includes transactionRequest; after allowance/permit handling, submit that transaction with the source-chain wallet.
    • If using POST /advanced/routes, first choose a route, then populate each step with POST /advanced/stepTransaction before execution.
    • EVM transaction requests usually include fields such as to, data, value, and gas fields.
    • Solana-originating transfers return Solana transaction data as base64 in transactionRequest.data; deserialize, sign, and send through the user's Solana wallet or SDK path.
    • Prefer SDK executeRoute for production multi-step execution because it manages allowance and balance checks, chain switching, transaction data retrieval, transaction submission, and status tracking.
    • Never mutate transactionRequest.data, calldata, recipient, refund, memo, or bridge-specific payloads after receiving them from LI.FI.
  6. Track status after source-chain submission

    • Poll /status every 10-30 seconds until terminal status.
    • Include fromChain, toChain, and bridge from the quote when available to speed up lookup.
    • Treat source-chain confirmation as only the start of a cross-chain transfer, not proof of final delivery.

Minimal Endpoint Set

Get a quote

Use for a single best route with transaction data included.

bash
curl --request GET \
  --url 'https://li.quest/v1/quote?fromChain=ARB&toChain=SOL&fromToken=0xaf88d065e77c8cC2239327C5EDb3A432268e5831&toToken=7vfCXTUXx5WJV5JADk17DUJ4ksgau7utNKj4b963voxs&fromAddress=YOUR_EVM_WALLET&toAddress=YOUR_SOL_WALLET&fromAmount=1000000000' \
  --header 'accept: application/json'

Required parameters:

  • fromChain: source chain ID or key.
  • toChain: destination chain ID or key.
  • fromToken: source token symbol or address.
  • toToken: destination token symbol or address.
  • fromAmount: amount in smallest unit.
  • fromAddress: sender wallet address.
  • toAddress: recipient wallet address when different from sender or when bridging across ecosystems.

Get multiple routes

Use when the user asks to compare routes or when the application needs route selection.

bash
curl --request POST \
  --url 'https://li.quest/v1/advanced/routes' \
  --header 'accept: application/json' \
  --header 'content-type: application/json' \
  --data '{
    "fromChainId": "ARB",
    "toChainId": "SOL",
    "fromTokenAddress": "0xaf88d065e77c8cC2239327C5EDb3A432268e5831",
    "toTokenAddress": "7vfCXTUXx5WJV5JADk17DUJ4ksgau7utNKj4b963voxs",
    "fromAmount": "1000000000",
    "fromAddress": "YOUR_EVM_WALLET",
    "toAddress": "YOUR_SOL_WALLET"
  }'

Check transfer status

bash
curl --request GET \
  --url 'https://li.quest/v1/status?txHash=SOURCE_TX_HASH&fromChain=ARB&toChain=SOL&bridge=BRIDGE_KEY' \
  --header 'accept: application/json'

Status handling:

  • NOT_FOUND: Transaction may not be indexed or mined yet. Retry with fromChain and bridge if known.
  • PENDING: Continue polling.
  • DONE + COMPLETED: Successful final delivery.
  • DONE + PARTIAL: Successful but output token may differ while preserving value semantics.
  • DONE + REFUNDED: Transfer failed but funds were refunded.
  • FAILED: Stop polling and surface the error/substatus.

List supported chains

bash
curl --request GET \
  --url 'https://li.quest/v1/chains?chainTypes=EVM,SVM' \
  --header 'accept: application/json'

Solana-only:

bash
curl --request GET \
  --url 'https://li.quest/v1/chains?chainTypes=SVM' \
  --header 'accept: application/json'

List Solana tokens

bash
curl --request GET \
  --url 'https://li.quest/v1/tokens?chains=SOL&chainTypes=SVM' \
  --header 'accept: application/json'

List bridges and exchanges

bash
curl --request GET \
  --url 'https://li.quest/v1/tools' \
  --header 'accept: application/json'

Solana-Specific Guidance

Use this section whenever either side of the transfer is Solana.

  • Use chainTypes=SVM when querying Solana chain support.
  • Use SOL as the Solana chain key in API examples.
  • Always pass a Solana toAddress when bridging from an EVM chain to Solana; the sender's EVM address is not a valid Solana recipient.
  • For SOL -> EVM transfers, expect transactionRequest.data to contain base64-encoded Solana transaction data. This is not EVM calldata.
  • Use a Solana wallet adapter or the LI.FI SDK to deserialize, sign, and submit Solana transactions. Do not manually rebuild the transaction unless the docs for the exact flow require it.
  • Validate token decimals from /tokens or /token; do not assume EVM and Solana versions of a token share decimals or addresses.
  • If status is slow, pass the source transaction hash plus fromChain, toChain, and bridge/tool key from the quote.

Execution Patterns

Direct API execution

Use direct API execution when you control the wallet/provider layer yourself. /quote returns executable transaction data immediately; /advanced/routes returns route choices and requires POST /advanced/stepTransaction for each selected step.

For EVM-originating quotes, a typical flow is:

  1. Check token allowance for quote.estimate.approvalAddress when the source token is not native.
  2. Submit an approval or use Permit/Permit2 if required by the selected flow.
  3. Send quote.transactionRequest through the EVM wallet.
  4. Poll /status with source transaction hash, quote.tool, source chain, and destination chain.

For Solana-originating quotes, do not treat transactionRequest.data as calldata. It is base64-encoded Solana transaction data; use the SDK or a Solana wallet-adapter/web3.js path to deserialize, sign, and submit it.

SDK execution

Prefer the SDK when building a frontend or when the agent needs full route execution rather than only quote/status lookup. The SDK exports createConfig, getQuote, getRoutes, executeRoute, EVM, Solana, and KeypairWalletAdapter from @lifi/sdk as of @lifi/sdk@3.16.3.

bash
npm install @lifi/sdk

Quote-only lookup:

typescript
import { createConfig, getQuote } from '@lifi/sdk'

createConfig({
  integrator: 'YourAppName',
  // Backend/trusted runtime only. Do not expose this in browsers.
  apiKey: process.env.LIFI_API_KEY,
})

const quote = await getQuote({
  fromChain: 'ARB',
  toChain: 'SOL',
  fromToken: '0xaf88d065e77c8cC2239327C5EDb3A432268e5831',
  toToken: '7vfCXTUXx5WJV5JADk17DUJ4ksgau7utNKj4b963voxs',
  fromAmount: '1000000000',
  fromAddress: 'YOUR_EVM_WALLET',
  toAddress: 'YOUR_SOL_WALLET',
})

console.log(quote.estimate.toAmount, quote.tool, quote.transactionRequest)

Route execution with SDK:

typescript
import { createConfig, EVM, executeRoute, getRoutes } from '@lifi/sdk'
import type { Chain } from 'viem'
import { createWalletClient, http } from 'viem'
import { privateKeyToAccount } from 'viem/accounts'
import { arbitrum, mainnet, optimism } from 'viem/chains'

const privateKey = process.env.PRIVATE_KEY as `0x${string}` // backend/testing only
const account = privateKeyToAccount(privateKey)
const chains = [arbitrum, mainnet, optimism]

const client = createWalletClient({
  account,
  chain: arbitrum,
  transport: http(),
})

createConfig({
  integrator: 'YourAppName',
  providers: [
    EVM({
      getWalletClient: async () => client,
      switchChain: async (chainId) =>
        createWalletClient({
          account,
          chain: chains.find((chain) => chain.id === chainId) as Chain,
          transport: http(),
        }),
    }),
  ],
})

const result = await getRoutes({
  fromChainId: 42161,
  toChainId: 10,
  fromTokenAddress: '0xaf88d065e77c8cC2239327C5EDb3A432268e5831',
  toTokenAddress: '0xDA10009cBd5D07dd0CeCc66161FC93D7c9000da1',
  fromAmount: '10000000',
  fromAddress: 'YOUR_EVM_WALLET',
})

const executedRoute = await executeRoute(result.routes[0], {
  updateRouteHook(route) {
    console.log(route)
  },
  async acceptExchangeRateUpdateHook() {
    // In a UI, ask the user before accepting.
    return false
  },
})

console.log(executedRoute)

Solana provider setup:

typescript
import { createConfig, KeypairWalletAdapter, Solana } from '@lifi/sdk'

const walletAdapter = new KeypairWalletAdapter('PRIVATE_KEY') // backend/testing only

createConfig({
  integrator: 'YourAppName',
  providers: [
    Solana({
      getWalletAdapter: async () => walletAdapter,
    }),
  ],
})

For user-facing Solana apps, use @solana/wallet-adapter-react and set Solana({ getWalletAdapter }) from the connected wallet adapter at runtime. Do not embed private keys in frontend code.

Error Handling

Handle both HTTP status codes and LI.FI API error codes.

Common API error codes:

  • 1002 NoQuoteError: No quote was found. Try another token, amount, destination, or route constraint.
  • 1005 RateLimitError: Back off and retry; use API key for higher limits if available.
  • 1007 SlippageError: Ask the user before increasing slippage.
  • 1008 ThirdPartyError: A bridge/exchange/tool failed. Retry later or request alternative routes.
  • 1009 TimeoutError: Retry with backoff or use route comparison to find another tool.
  • 1011 ValidationError: Check chain IDs, token addresses, amount units, and address format.
  • 1012 RpcFailure: Retry later or configure a more reliable RPC through the SDK when applicable.

Common tool error codes:

  • NO_POSSIBLE_ROUTE: No supported path for the requested action.
  • INSUFFICIENT_LIQUIDITY: Reduce amount or choose another token/path.
  • TOOL_TIMEOUT: Retry or exclude the timing-out tool if the API request supports tool constraints.
  • AMOUNT_TOO_LOW / AMOUNT_TOO_HIGH: Adjust amount.
  • FEES_HIGHER_THAN_AMOUNT: Amount is uneconomic; increase transfer size or choose another route.
  • DIFFERENT_RECIPIENT_NOT_SUPPORTED: Use the same sender/recipient where required or choose another tool.
  • CANNOT_GUARANTEE_MIN_AMOUNT: Explain the risk and ask the user before proceeding.

Guidelines

  • DO query live chains, tokens, and tools before giving integration advice.
  • DO use /quote for simple execution and /advanced/routes for comparison.
  • DO show estimated output, fees, bridge/exchange tool, recipient, and slippage before signing.
  • DO poll /status for cross-chain transfers until a terminal state.
  • DO preserve LI.FI transaction data exactly as returned.
  • DO register in the LI.FI Partner Portal for API keys when higher rate limits are needed, and keep keys server-side.
  • DO NOT hardcode stale bridge lists, token lists, or chain support.
  • DO NOT treat source-chain confirmation as final cross-chain completion.
  • DO NOT increase slippage, change recipients, or alter transaction data without explicit user consent.
  • DO NOT store or request private keys. Use wallet signing flows.
  • DO NOT present the Widget as the only integration path when the API or SDK fits better.

Common Errors

Error: No route found

Cause: Unsupported chain/token pair, insufficient liquidity, amount too small or too large, or restrictive route options.

Solution: Verify /chains, /tokens, and /tools; try canonical token addresses; adjust amount; request /advanced/routes for alternatives.

Error: Solana transaction data treated like EVM calldata

Cause: A SOL-originating quote returns base64 Solana transaction data in transactionRequest.data, not an EVM to/data/value call.

Solution: Use a Solana wallet/SDK path to deserialize, sign, and submit the transaction. Do not send it through an EVM provider.

Error: Status endpoint returns NOT_FOUND

Cause: Transaction not indexed yet, wrong source transaction hash, missing fromChain, or wrong bridge/tool filter.

Solution: Retry with backoff and pass fromChain, toChain, and bridge from the quote. Confirm the hash is the source-chain transaction hash.

Error: Wrong amount due to decimals

Cause: The amount was not converted into the source token's smallest unit, or token decimals were assumed incorrectly across ecosystems.

Solution: Fetch token metadata from /tokens or /token and convert amounts before requesting a quote.

Error: Different recipient unsupported

Cause: Some tools do not support sending to a recipient different from the sender.

Solution: Request another route/tool or use the same recipient when acceptable to the user.

References

Frequently asked questions

What does the Lifi AI skill do?

Integrate LI.FI for cross-chain swaps, bridging, payments, route discovery, and transfer status tracking across Solana, EVM, Bitcoin, and Sui. Use when building Solana applications or AI agents that need quotes, routes, executable transactions, supported chains/tokens/tools, or cross-chain transfer monitoring.

Why use Lifi on TypingMind?

Because you install it once and use it with any model. Lifi is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Lifi in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/sendaifun/skills/tree/main/skills/lifi. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Lifi?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Lifi?

As many as you like. As long as a model supports skills, you can use Lifi with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Lifi AI skill free?

Yes. It is published on GitHub by sendaifun under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇