Pump Swarm logo

Pump Swarm

CommunityPopular
alsk1992
pump-swarm

Coordinated multi-wallet trading on Pump.fun

Overview

Publisheralsk1992
RepositoryCloddsBot
Skill namepump-swarm
Stars
3K
Forks
353
Bundled files
1
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.

  • 1 bundled files

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

  • Open source

    Published by alsk1992 on GitHub. Read the source before you install it.

Installation

Install the Pump Swarm 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/alsk1992/CloddsBot/tree/main/src/skills/bundled/pump-swarm
  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/alsk1992/CloddsBot.git /tmp/CloddsBot
mkdir -p .claude/skills
cp -r /tmp/CloddsBot/src/skills/bundled/pump-swarm .claude/skills/pump-swarm
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pump Swarm 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 Pump Swarm 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 Pump Swarm 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.

Pump.fun Swarm Trading

Coordinate up to 20 wallets to execute synchronized trades on Pump.fun tokens.

Quick Start

bash
# Set up wallets
export SOLANA_PRIVATE_KEY="your-main-wallet-key"     # wallet_0
export SOLANA_SWARM_KEY_1="second-wallet-key"        # wallet_1
export SOLANA_SWARM_KEY_2="third-wallet-key"         # wallet_2
# ... up to SOLANA_SWARM_KEY_20

# Optional
export SOLANA_RPC_URL="https://your-rpc.com"

Commands

Wallet Management

/swarm wallets              List all swarm wallets with addresses
/swarm balances             Fetch SOL balances from chain
/swarm enable <wallet_id>   Enable a wallet for trading
/swarm disable <wallet_id>  Disable a wallet

Trading

/swarm buy <mint> <sol> [options]      Buy with all enabled wallets
/swarm sell <mint> <amount|%> [opts]   Sell from wallets with positions

Position Management

/swarm position <mint>      Show cached token positions
/swarm refresh <mint>       Fetch fresh positions from chain (required before sell)

Execution Modes

ModeFlagBest ForDescription
Parallel--parallelSpeed (>5 wallets)All wallets execute simultaneously
Bundle--bundleAtomicity (≤5)Single Jito bundle, all-or-nothing
Multi-Bundle--multi-bundleAtomicity (6-20)Multiple Jito bundles in parallel
Sequential--sequentialStealthStaggered 200-400ms delays

Auto Mode Selection (Default)

  • 1 wallet → Parallel (direct submit)
  • 2-5 wallets → Single Jito Bundle
  • 6-20 wallets → Multi-Bundle (chunks of 5)

Other Options

OptionDescription
--preset <name>Apply a saved preset
--wallets <id1,id2>Use specific wallets only
--slippage <bps>Slippage tolerance (default: 500 = 5%)
--pool <pool>Pool: pump, raydium, auto (pumpfun only)
--dex <dex>DEX: pumpfun (default), bags, meteora
--pool-address <addr>Specific pool address (for Meteora)

Examples

bash
# Buy 0.1 SOL worth on each enabled wallet (auto mode)
/swarm buy ABC123mint... 0.1

# Buy with specific wallets only
/swarm buy ABC123mint... 0.2 --wallets wallet_0,wallet_1

# Buy with a preset
/swarm buy ABC123mint... 0.1 --preset stealth

# Sell 100% with multiple Jito bundles (for >5 wallets)
/swarm sell ABC123mint... 100% --multi-bundle

# Sell 50% with staggered timing (stealth mode)
/swarm sell ABC123mint... 50% --sequential

# Sell with preset
/swarm sell ABC123mint... 100% --preset fast

# Check positions before selling
/swarm refresh ABC123mint...
/swarm position ABC123mint...

# Multi-DEX examples
/swarm buy ABC123mint... 0.1 --dex bags          # Buy on Bags.fm
/swarm buy ABC123mint... 0.1 --dex meteora       # Buy on Meteora DLMM
/swarm sell ABC123mint... 100% --dex bags        # Sell on Bags.fm

Execution Modes Deep Dive

Parallel (Default for >5 wallets)

  • Speed: All wallets submit simultaneously via Promise.all
  • Risk: No atomicity - some may succeed, others fail
  • Use when: Speed is priority, or bundles keep failing

Jito Bundle (Default for 2-5 wallets)

  • Atomic: All transactions succeed or all fail together
  • MEV-protected: No front-running between your own wallets
  • Cost: ~10,000 lamports tip per bundle
  • Limit: Max 5 transactions per bundle (Jito constraint)

Multi-Bundle (Recommended for 6-20 wallets)

  • Chunked: Splits wallets into groups of 5
  • Parallel bundles: All chunks submit simultaneously
  • Partial atomicity: Each chunk is atomic, but chunks are independent
  • Example: 12 wallets → 3 bundles of [5, 5, 2] wallets

Sequential (Stealth mode)

  • Staggered: 200-400ms random delay between wallets
  • Amount variance: ±5% to avoid detection patterns
  • Rate limited: 5 seconds minimum between trades per wallet
  • Use when: Want to avoid pattern detection

How It Works

Buy Flow

  1. Refreshes SOL balances from chain
  2. Filters wallets with sufficient balance (≥0.01 SOL + amount)
  3. Builds transaction for each wallet locally via the official @pump-fun/pump-sdk (or @pump-fun/pump-swap-sdk for graduated tokens)
  4. Signs all transactions locally
  5. Submits via selected execution mode
  6. Reports results per wallet

Sell Flow

  1. Fetches actual token balances from chain (critical!)
  2. Filters wallets with positions
  3. Calculates sell amount (% of position or exact)
  4. Builds and signs transactions
  5. Submits via selected execution mode
  6. Reports results per wallet

Safety Features

  • Balance check: Verifies sufficient SOL before buy
  • Position check: Fetches real token balances before sell
  • Max amount: Rejects buy amounts > 10 SOL per wallet
  • Confirmation timeout: 60 second timeout per transaction
  • Error reporting: Shows detailed errors per wallet
  • Bundle fallback: Failed bundles automatically retry as parallel

Configuration

Env VariableDescription
SOLANA_PRIVATE_KEYMain wallet (wallet_0)
SOLANA_SWARM_KEY_1..20Additional swarm wallets
SOLANA_RPC_URLCustom RPC endpoint (faster = better)
BAGS_API_KEYBags.fm API key (required for bags DEX)

Multi-DEX Support

The swarm system supports trading across multiple DEXes:

DEXFlagBest ForRequires
Pump.fun (bonding curve)--dex pumpfun (default)Memecoins, new launches-
Pump.fun (PumpSwap, graduated)--dex pumpswapTokens that migrated off the bonding curve-
Bags.fm--dex bagsBags-launched tokensBAGS_API_KEY
Meteora--dex meteoraDLMM pools, concentrated liquidity-

Multi-DEX Examples

bash
# Buy on Pump.fun (default)
/swarm buy ABC123... 0.1

# Buy on Bags.fm
/swarm buy ABC123... 0.1 --dex bags

# Buy on Meteora with specific pool
/swarm buy ABC123... 0.1 --dex meteora --pool-address <pool_address>

# Sell on Bags.fm with stealth preset
/swarm sell ABC123... 100% --dex bags --preset stealth

Notes

  • Default DEX is pumpfun for backward compatibility
  • Bags requires BAGS_API_KEY - will error if missing
  • Meteora can auto-discover pools or use a specific --pool-address
  • All execution modes (parallel, bundle, sequential) work with all DEXes
  • Presets work with all DEXes

Agent Tools (12)

ToolDescription
swarm_walletsList all swarm wallets
swarm_balancesRefresh SOL balances from chain
swarm_buyCoordinated buy across wallets (supports preset param)
swarm_sellCoordinated sell across wallets (supports preset param)
swarm_positionGet cached positions
swarm_refreshFetch fresh positions from chain
swarm_enableEnable a wallet
swarm_disableDisable a wallet
swarm_preset_saveSave a trading preset
swarm_preset_listList saved presets
swarm_preset_getGet preset details
swarm_preset_deleteDelete a preset

Scaling Notes

  • Max wallets: 20 (wallet_0 + SOLANA_SWARM_KEY_1..20)
  • Jito bundle limit: 5 txs per bundle (handled automatically)
  • Multi-bundle parallel: All bundles submit simultaneously
  • RPC recommendation: Use dedicated RPC for best performance
  • Tip amount: 10,000 lamports per bundle (~$0.002)

Presets

Save and reuse trading configurations across tokens and strategies.

Preset Commands

bash
/swarm preset save <name> [options]   Save a preset
/swarm preset list [type]             List presets
/swarm preset show <name>             Show preset details
/swarm preset delete <name>           Delete a preset

Preset Save Options

OptionDescription
--type <type>Preset type: strategy, token, wallet_group
--desc "..."Preset description
--mint <addr>Token address (for token presets)
--amount <sol>Default SOL per wallet
--slippage <bps>Slippage in basis points
--pool <pool>Pool: pump, raydium, auto
--mode <mode>parallel, bundle, multi-bundle, sequential
--wallets <ids>Wallet IDs (for wallet_group presets)

Built-in Presets

NameModeSlippagePoolUse Case
fastparallel5%autoSpeed priority
atomicmulti-bundle5%autoAll-or-nothing execution
stealthsequential3%autoPattern avoidance
aggressiveparallel10%pumpHigh volatility tokens
safebundle2%autoConservative trading

Preset Examples

bash
# Create a custom strategy preset
/swarm preset save my_stealth --type strategy --mode sequential --slippage 300

# Create a token preset for BONK
/swarm preset save bonk_entry --type token --mint DezXAZ... --slippage 1000 --amount 0.1

# Create a wallet group preset
/swarm preset save top5 --type wallet_group --wallets wallet_0,wallet_1,wallet_2,wallet_3,wallet_4

# List all presets
/swarm preset list

# List only strategy presets
/swarm preset list strategy

# Use preset in trade
/swarm buy ABC... 0.1 --preset my_stealth
/swarm sell ABC... 100% --preset fast

# Delete a preset
/swarm preset delete old_preset

Preset Types

TypePurpose
strategyReusable trading settings (mode, slippage, pool)
tokenToken-specific settings with saved mint address
wallet_groupNamed wallet combinations for specific trades

Troubleshooting

IssueSolution
Bundle keeps failingTry --parallel or check network congestion
Positions not showingRun /swarm refresh <mint> first
Insufficient balanceCheck with /swarm balances
Slow executionUse dedicated RPC via SOLANA_RPC_URL
Some wallets skippedWallet disabled or insufficient balance
Preset not foundCheck name with /swarm preset list

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 Pump Swarm AI skill do?

Coordinated multi-wallet trading on Pump.fun

Why use Pump Swarm on TypingMind?

Because you install it once and use it with any model. Pump Swarm 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 Pump Swarm in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/alsk1992/CloddsBot/tree/main/src/skills/bundled/pump-swarm. 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 Pump Swarm?

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 Pump Swarm?

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

Is the Pump Swarm AI skill free?

Yes. It is published on GitHub by alsk1992 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 👇