Dex Execution logo

Dex Execution

Community
agiprolabs
dex-execution

Solana DEX swap execution via Jupiter aggregator including quoting, transaction building, signing, and confirmation

Overview

Publisheragiprolabs
Repositoryclaude-trading-skills
Skill namedex-execution
Stars
434
Forks
85
Bundled files
5
LicenseMIT
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.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

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

Installation

Install the Dex Execution 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.
    https://github.com/agiprolabs/claude-trading-skills/tree/main/skills/dex-execution
  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/agiprolabs/claude-trading-skills.git /tmp/claude-trading-skills
mkdir -p .claude/skills
cp -r /tmp/claude-trading-skills/skills/dex-execution .claude/skills/dex-execution
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Dex Execution 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 Dex Execution 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 Dex Execution 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.

DEX Execution — Solana Swap Execution via Jupiter

Execute token swaps on Solana through Jupiter, the dominant DEX aggregator routing across Raydium, Orca, Meteora, Phoenix, Lifinity, and 20+ other venues.

Overview

Jupiter aggregates liquidity across all major Solana DEXes to find optimal swap routes. A single swap may split across multiple pools and hop through intermediate tokens to minimize price impact. The Jupiter v6 API handles route discovery, transaction building, and fee optimization — your code handles quoting, user confirmation, signing, and submission.

Base URL: https://quote-api.jup.ag/v6

Execution Pipeline

Every swap follows this seven-step pipeline. Never skip steps 2-3 (display and confirm).

Step 1 — Get Quote

python
import httpx

params = {
    "inputMint": "So11111111111111111111111111111111111111112",   # SOL
    "outputMint": "EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v", # USDC
    "amount": 1_000_000_000,  # 1 SOL in lamports
    "slippageBps": 50,        # 0.5%
}
resp = httpx.get("https://quote-api.jup.ag/v6/quote", params=params)
quote = resp.json()

Step 2 — Display Quote to User

Always show these fields before proceeding:

FieldSource
Input amountquote["inAmount"] (in token decimals)
Output amountquote["outAmount"]
Minimum receivedquote["otherAmountThreshold"]
Price impactquote["priceImpactPct"]
Routequote["routePlan"] — DEXes used
SlippageThe slippageBps you requested

Step 3 — Require User Confirmation

⚠️  SWAP PREVIEW
    Selling:   1.000 SOL
    Buying:    ~142.50 USDC
    Min recv:  141.79 USDC (0.5% slippage)
    Impact:    0.01%
    Route:     Raydium V4 → USDC

    Proceed? [y/N]

NEVER proceed without explicit "yes" from the user.

Step 4 — Build Transaction

python
swap_body = {
    "quoteResponse": quote,
    "userPublicKey": "YourPubkeyBase58...",
    "wrapAndUnwrapSol": True,
    "dynamicComputeUnitLimit": True,
    "prioritizationFeeLamports": "auto",
}
resp = httpx.post("https://quote-api.jup.ag/v6/swap", json=swap_body)
swap_data = resp.json()
swap_tx = swap_data["swapTransaction"]  # base64-encoded transaction

Step 5 — Sign Transaction

python
import base64
from solders.transaction import VersionedTransaction
from solders.keypair import Keypair

raw_tx = base64.b64decode(swap_tx)
tx = VersionedTransaction.from_bytes(raw_tx)
keypair = Keypair.from_base58_string(os.environ["WALLET_PRIVATE_KEY"])
tx.sign([keypair])

Step 6 — Submit Transaction

python
rpc_url = os.environ.get("SOLANA_RPC_URL", "https://api.mainnet-beta.solana.com")
signed_bytes = bytes(tx)
payload = {
    "jsonrpc": "2.0", "id": 1,
    "method": "sendTransaction",
    "params": [
        base64.b64encode(signed_bytes).decode(),
        {"encoding": "base64", "skipPreflight": False,
         "maxRetries": 3, "preflightCommitment": "confirmed"}
    ],
}
resp = httpx.post(rpc_url, json=payload)
sig = resp.json()["result"]
print(f"Submitted: https://solscan.io/tx/{sig}")

Step 7 — Confirm Transaction

python
import time

for attempt in range(30):
    payload = {
        "jsonrpc": "2.0", "id": 1,
        "method": "getSignatureStatuses",
        "params": [[sig], {"searchTransactionHistory": False}],
    }
    resp = httpx.post(rpc_url, json=payload)
    status = resp.json()["result"]["value"][0]
    if status and status.get("confirmationStatus") in ("confirmed", "finalized"):
        print(f"Confirmed at slot {status['slot']}")
        break
    time.sleep(2)
else:
    print("Transaction not confirmed within 60s — check explorer")

Jupiter API v6 Endpoints

EndpointMethodPurpose
/quoteGETGet best-price quote with routing
/swapPOSTBuild a swap transaction from a quote
/swap-instructionsPOSTGet individual instructions (advanced)
/price?ids=token1,token2GETSimple price lookup (v2)
/tokensGETList all supported tokens

See references/jupiter_api.md for full parameter and response documentation.

Key Parameters

Slippage (slippageBps)

Token TypeRecommended RangeNotes
SOL, USDC, major tokens50-100 (0.5-1%)Stable liquidity
Mid-cap tokens100-300 (1-3%)Variable liquidity
PumpFun / meme tokens500-2000 (5-20%)Thin books, high volatility
New launches (<1h old)1000-3000 (10-30%)Extreme volatility

Dynamic Slippage

Set dynamicSlippage: true in the swap request to let Jupiter auto-adjust slippage based on current market conditions. Preferred for most use cases.

Priority Fees (prioritizationFeeLamports)

Priority fees determine transaction ordering within a block.

LevelmicroLamportsWhen to Use
Low10,000-50,000Normal conditions
Medium50,000-200,000Moderate congestion
High200,000-1,000,000High congestion / time-sensitive
Urgent1,000,000-5,000,000Meme coin launches, NFT mints
Auto"auto"Jupiter estimates for you

Use "auto" for most cases. For fine-grained control, query Helius getPriorityFeeEstimate.

Other Parameters

  • onlyDirectRoutes: Skip multi-hop routing. Faster but may get worse price.
  • asLegacyTransaction: Use legacy format instead of versioned transactions. Required for some older wallets.
  • maxAccounts: Limit accounts in transaction (default 64). Lower values reduce route options but improve confirmation reliability.
  • platformFeeBps: Integrator fee in basis points. Taken from output amount.
  • wrapAndUnwrapSol: Auto wrap/unwrap SOL ↔ wSOL (default true).
  • dynamicComputeUnitLimit: Auto-set compute budget based on simulation.

Priority Fee Estimation

python
# Using Helius getPriorityFeeEstimate
helius_url = f"https://mainnet.helius-rpc.com/?api-key={HELIUS_KEY}"
payload = {
    "jsonrpc": "2.0", "id": 1,
    "method": "getPriorityFeeEstimate",
    "params": [{"accountKeys": [input_mint, output_mint],
                "options": {"recommended": True}}],
}
resp = httpx.post(helius_url, json=payload)
fee = resp.json()["result"]["priorityFeeEstimate"]

Transaction Confirmation Strategy

See references/transaction_lifecycle.md for the full Solana transaction lifecycle.

  1. Submit with skipPreflight: False to catch obvious errors
  2. Poll getSignatureStatuses every 2 seconds for up to 60 seconds
  3. If not confirmed: rebuild transaction with fresh blockhash and retry (max 3 attempts)
  4. Blockhash expiry: transactions expire ~60 seconds after blockhash was fetched
  5. Final check: verify token balance changed as expected

Error Handling

ErrorCauseRecovery
"Slippage tolerance exceeded"Price moved beyond slippageBpsIncrease slippage or retry
"Insufficient funds"Not enough input token or SOL for feesCheck balance before quoting
"Transaction expired"Blockhash too oldRebuild with fresh blockhash
"Transaction simulation failed"Various — check logsParse simulation logs for root cause
"Too many accounts"Route uses too many accountsSet maxAccounts lower or use onlyDirectRoutes
HTTP 429Rate limitedBack off and retry with exponential delay
HTTP 400 "No route found"No liquidity path existsCheck token mints are correct, try larger slippage

Safety Requirements

These are non-negotiable requirements for any execution code.

  1. ALWAYS show quote details (amounts, price impact, route) before execution
  2. ALWAYS require explicit user confirmation — never auto-execute
  3. Default to simulation mode — do not sign or submit unless explicitly enabled
  4. Never store or log private keys — load from env vars, use immediately, discard
  5. Never use 100% slippage — this is a common scam/exploit vector
  6. Verify token addresses — confirm mints match expected tokens before swapping
  7. Check price impact — warn if >2%, block if >10% unless user overrides
  8. Maintain SOL reserve — keep 0.05 SOL minimum for rent and future fees

See references/safety_checklist.md for the complete pre/during/post execution checklist.

Integration with Other Skills

SkillIntegration
slippage-modelingEstimate optimal slippageBps based on token liquidity profile
liquidity-analysisVerify pool depth supports trade size before quoting
position-sizingCalculate trade amount based on risk parameters
risk-managementEnforce portfolio-level exposure limits before execution
jupiter-apiUnderlying API documentation for Jupiter endpoints
helius-apiPriority fee estimation and transaction monitoring

Files

References

  • references/jupiter_api.md — Complete Jupiter v6 API parameter and response reference
  • references/transaction_lifecycle.md — Solana transaction lifecycle, priority fees, retry strategies
  • references/safety_checklist.md — Pre/during/post execution verification checklist

Scripts

  • scripts/get_quote.py — Fetch and display Jupiter swap quotes with route analysis
  • scripts/simulate_swap.py — Build and simulate swap transactions without submitting

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Dex Execution AI skill do?

Solana DEX swap execution via Jupiter aggregator including quoting, transaction building, signing, and confirmation

Why use Dex Execution on TypingMind?

Because you install it once and use it with any model. Dex Execution 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 Dex Execution in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/agiprolabs/claude-trading-skills/tree/main/skills/dex-execution. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Dex Execution?

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 Dex Execution?

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

Is the Dex Execution AI skill free?

Yes. It is published on GitHub by agiprolabs under the MIT 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 👇