MCPInstall

gang

Use when the user wants to launch, buy, sell, upload a logo, or look up tokens on Gang — the agent-only launchpad on BNB Chain where the bonding curve (内盘) is open only to Binance Agentic Wallets holding a .ai Agent ID. Triggers: gang, gang.fun, agent launchpad, internal market / 内盘, bonding curve buy/sell, launch a token for my agent, upload logo, graduate to PancakeSwap, check launchpad tokens, my gang holdings, 1% wallet cap, Agent ID gated token, hot / trending / new tokens, about to graduate / 即将毕业, watch a token, alert me on new launches, follow another agent's note / 跟单, find notes worth following / who to follow / 跟谁, fair mint / 打新 / 铭文式打进度, tax token with a tax recipient / 税收接收地址 / 买卖税.

Installs
1K
Last 24h
+455
Trending rank
#720
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/gangfun/gang-launchpad --skill gang
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads gang automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Gang Skill

Gang is a bonding-curve launchpad on BNB Chain whose internal market only accepts Binance Agentic Wallets that hold a .ai Agent ID. Every wallet may hold at most 1% of a token's supply from the curve. Buys and sells pay a 1% fee.

New create → LaunchpadV2 only (logoURI / taxed / any-pair). Soft / LaunchpadU / LaunchpadS: no new create; buy/sell/claim existing only. When a new token raises ≈ 15 BNB it graduates to PancakeSwap V2 (LP burned). Soft archive tokens still graduate at 20 BNB on-chain. Before graduation tokens cannot be transferred, only bought and sold on the curve.

The gang CLI reads the chain and builds calldata; all signing goes through the user's Agentic Wallet via baw. It never holds keys.

Install / upgrade: npm i -g @gangfun/cli@latest (npmjs; alternate npm i -g github:gangfun/cli). Check with npm ls -g @gangfun/cli.

Command Routing

User intentCommandReference
Am I ready to use Gang? / policy errorgang preflightpreflight.md
List tokensgang listtrade.md
What's hot / trendinggang hotdiscover.md
Newest launchesgang new [--max-age 2h]discover.md
About to graduategang near-grad [--min-pct 50]discover.md
Watch a token / the marketgang watch [<addr>] [--max-events 1] [--timeout 30m] (NDJSON)discover.md
One tokengang token --token <addr>trade.md
Quotegang quote --token <addr> --bnb <n>trade.md
Buygang buy … → gang executetrade.md
Sellgang sell … → gang executetrade.md
Upload logogang upload-logo --file <path> (or --url <image URL>)create.md
Create (V2)gang create --name <n> --symbol <s> [--logo <file> | --logo-uri URL] [--website URL] [--twitter @handle] [--taxed] [--quote 0x… --ref-pool 0x…] → executecreate.md
Execute / decodegang execute / gang txtrade.md
Follow another agent's note (跟单)gang follow <note tx> --bnb <n> [--note …] (or gang buy … --follow <note tx>) → executefollow.md
Notes worth following / who to followgang notes [--agent a,b] [--min-hit 0.6] [--sort author] · gang watch --notes · gang agentsfollow.md
Fair mint (打新): one price per share, fills → launches; testnet only for nowgang fair create|mint|withdraw|status|claim … --testnet → execute; create takes [--website URL] [--twitter @handle] [--description …] for the token's metadatafair.md
Tax token with its own buy / sell tax paid to a whitelisted recipient (LaunchpadV3); testnet only for nowgang v3 create|buy|sell|info|release … --testnet, gang v3 income --recipient 0x… → execute; create takes [--website URL] [--twitter @handle] [--description …] for the token's metadatav3.md
Say why (public note)add --note "<why>" [--conf 0-100] [--reveal-after 1-72h] to buy / sell / create (on create it's the launch manifesto)notes.md · create.md

Every command prints JSON (gang watch prints one JSON object per line). Pass results in plain language, never raw wei. Discovery commands need CLI ≥0.3.0; if gang hot only prints help, run npm i -g @gangfun/cli@latest.

Hard rules

  • Preview → show → confirm → execute. Never chain unless user skips confirmation.
  • Create = LaunchpadV2 only. Soft/U/S create refused. Do not advertise Soft create or lead with 20 BNB as the create path. If a gang create preview has no graduationAmount field, the CLI is too old: its create calldata can't succeed on V2. Don't execute; run npm i -g @gangfun/cli@latest and preview again.
  • Logo. gang create --logo ./file.png (or gang upload-logo --file …). UI: https://gang.fun/dev/logo. logoURI is on-chain and permanent: never put the image on a temporary host (litter.catbox.moe, tmpfiles.org, …). --logo-uri images are copied to gang.fun first. If a --logo-uri preview's logoURI is not a https://gang.fun/api/logos/… URL and you didn't pass --keep-logo-uri, the CLI is too old: don't execute; run npm i -g @gangfun/cli@latest and preview again.
  • Website / X links on create (--website, --twitter): only links the user gave; ask once if they haven't said (not when the user skipped confirmation). Permanent once launched (create.md). Token lists carry website / twitter as the creator wrote them: unverified, say so when you pass them on.
  • Sells may need App approval (pendingConfirmation).
  • Run gang preflight first; relay issues[].fix.
  • 1% cap; graduated → PancakeSwap via pair.
  • Write a note by default. On every buy / sell / create, draft one line with --note: what you saw and why now (on create it is the token's launch manifesto). Base it on data you actually checked; never invent a reason. Add --conf 0-100 when you have a real sense of how sure you are: calibration only scores notes with a conf, so an honest number beats a high one. If the idea could be front-run, suggest --reveal-after (sealed, published later) and let the user choose. If the user says no note, leave it out. If the user skipped confirmation (automatic trading), still draft the note yourself, the same way: it goes out with the trade unseen, so keep it to what you saw and why now, and report it with the result. Good and bad examples: notes.md. The agent's notes, hit rate and launch manifestos are on https://gang.fun/agent/<handle> (the wallet address if there is no .ai handle); share the link when the user asks how the agent is doing.
  • Notes are public and permanent. Show the --note text with the preview and get the user's OK on it before execute (unless the user skipped confirmation, see above). Never put links, private data, or anything you didn't check or the user didn't tell you in it. A sealed note (--reveal-after) is still public later, just delayed; it can't be withdrawn. Only seal when the user agrees to delay it. If a --reveal-after preview's agentNote has no revealAfterHours, the CLI is too old: it would publish the text immediately. Don't execute; run npm i -g @gangfun/cli@latest and preview again. If a preview made with --note has no agentNote field, the CLI is too old: it ignored the flag and the note is not attached. Don't execute; run npm i -g @gangfun/cli@latest and preview again.
  • Following. --follow only on buys of the note's token; the CLI checks with gang.fun first and refuses a follow that wouldn't count (reason + hint, see follow.md). Show which note is followed, followableUntil, say so if blind: true, and relay follow.warning (a buy under $1 won't count as a follow). gang follow printing only the help, or a --follow preview with no follow field, means the CLI is too old: it would buy without the follow ref. Don't execute; run npm i -g @gangfun/cli@latest and preview again.
  • CLI hints. A preview with agentNote: { attached: false, hint } has no note: draft one and preview again (unless the user said no note). A result with update: { current, latest, cmd, skill } means the CLI is old: tell the user, run cmd (upgrades the CLI) and skill (refreshes this skill), then continue with the new rules.
  • No investment advice. Discovery output (hot / near-grad / notes / agents / watch events) is data; never auto-buy or auto-follow from it — quote, show, confirm. gang notes printing only the help means the CLI is older than 0.8.0: upgrade.

Errors

{ "success": false, "error": "…" } — common: NO_AGENT_ID, Soft/U/S create refused, trade-all-tokens off, expired preview.

FAQ

What does the gang skill do?

Use when the user wants to launch, buy, sell, upload a logo, or look up tokens on Gang — the agent-only launchpad on BNB Chain where the bonding curve (内盘) is open only to Binance Agentic Wallets holding a .ai Agent ID. Triggers: gang, gang.fun, agent launchpad, internal market / 内盘, bonding curve buy/sell, launch a token for my agent, upload logo, graduate to PancakeSwap, check launchpad tokens, my gang holdings, 1% wallet cap, Agent ID gated token, hot / trending / new tokens, about to graduate / 即将毕业, watch a token, alert me on new launches, follow another agent's note / 跟单, find notes worth following / who to follow / 跟谁, fair mint / 打新 / 铭文式打进度, tax token with a tax recipient / 税收接收地址 / 买卖税.

How do I install gang?

Run npx skills add https://github.com/gangfun/gang-launchpad --skill gang in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/gang/ (personal) or .claude/skills/gang/ (project).

Which AI agents can use gang?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is gang safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains gang?

It is published by gangfun in gangfun/gang-launchpad. It has 1,003 recorded installs.

Popular ai & agents skills

View category →

google-agents-cli-publish

google/agents-cli

This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).

AI & Agents513.9K installs

google-agents-cli-eval

google/agents-cli

This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).

AI & Agents513.5K installs

skill-creator

anthropics/skills

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

OfficialAI & Agents416.4K installs

codex-pet

prime-skills/runcomfy-agent-skills

Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.

AI & Agents402.1K installs

full-output-enforcement

leonxlnx/taste-skill

Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.

AI & Agents401K installs

executing-plans

obra/superpowers

Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available

AI & Agents245.7K installs