agentleFS
ClawpumpClawpump/agents-skills/skills/rug-check/SKILL.md

rug-check

Skill3 starsChanged 7 months agoMITView on GitHub

Solana token safety diligence: mint/freeze authority, LP lock/burn, top-holder concentration, RugCheck score. Read-only — run before swapping into or launching alongside any token.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read-only safety diligence for any Solana token mint.

    line 25
  2. Distinctive:

    Before launching or buying alongside an existing token.

    line 41
  3. Very specific:

    Always show percentages and the score band.

    line 90
  4. Very specific:

    Use the terminal tool with python3/curl (not web_extract).

    line 114
  5. Very specific:

    Read-only: this skill complements — never replaces — the ClawPump MCP (mcp_clawpump_*) trade and launch tools.

    line 116

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

rug-check — Solana Token Safety

Read-only safety diligence for any Solana token mint. Never signs anything. It pairs with the ClawPump MCP trade/launch tools (swap_execute, launch_token, …) as the pre-flight that decides whether a token is safe to touch.

Two independent sources, cross-checked:

  1. On-chain (authoritative): Solana JSON-RPC for mint/freeze authority and real top-holder concentration — facts, not opinions.
  2. RugCheck (heuristic): a community risk score + flagged risks.

Helper: scripts/rugcheck.py. Endpoint reference: references/api-endpoints.md.

When to Use

  • Before any swap_execute into an unfamiliar token.
  • A user pastes a mint and asks "is this safe / is it a rug / can I ape this?"
  • Before launching or buying alongside an existing token.
  • Vetting a token a tracked wallet/whale just bought.

Quick Check (do this first)

python3 ~/.hermes/skills/crypto/rug-check/scripts/rugcheck.py check <MINT>

This prints a consolidated verdict: authority status, top-10 concentration, RugCheck score, and a CLEAN / CAUTION / RUGGABLE line with the reasons.

Key Concepts

  • Mint authority — if NOT revoked (non-null), the team can mint unlimited new supply and dilute you. Revoked (null) is the safe state.
  • Freeze authority — if NOT revoked, the team can freeze your tokens so you can't sell. Revoked (null) is safe.
  • LP locked vs burned — liquidity that's burned/locked can't be pulled (the classic "rug"). Unlocked LP is a red flag.
  • Top-holder concentration — a few wallets holding most of the supply can dump on everyone. >30% top-10 is elevated, >60% is dangerous. (Some large holders are locked LP / CEX / vesting — verify owners before concluding.)
  • RugCheck score — heuristic, not a guarantee. Use it as a signal, not a verdict.

Individual Commands

# RPC authority cross-check (mint can't be inflated / frozen?)
python3 scripts/rugcheck.py authority <MINT>

# Holder concentration from on-chain largest accounts
python3 scripts/rugcheck.py holders <MINT> --top 10

# RugCheck quick summary (+ risk flags) / full report
python3 scripts/rugcheck.py summary <MINT>
python3 scripts/rugcheck.py report  <MINT>

Raw curl equivalents (no key needed) are in references/api-endpoints.md.

Presenting Results

Lead with the verdict line, then the evidence:

⚠️ CAUTION — mint authority is still ACTIVE (supply can be inflated) and the top 10 wallets hold 41%. RugCheck score 1,800. Mint/freeze: mint ACTIVE 🔴 / freeze revoked ✅. I'd size small or skip.

Always show percentages and the score band. Never auto-execute a trade off a "clean" result — confirm with the user (and follow the ClawPump confirmation rules) before any swap_execute / launch.

Rate Limits & Keys

  • RugCheck public endpoints are rate-limited (best-effort; brand-new tokens may have no report yet — fall back to the on-chain authority/holder check).
  • Solana RPC: defaults to the public api.mainnet-beta.solana.com. For heavier use, set SOLANA_RPC_URL or HELIUS_API_KEY (free at dev.helius.xyz) — the script picks them up automatically.

Pitfalls

  • A "clean" mint can still dump — LP unlocked, insider wallets, or social manipulation aren't fully captured here. This is risk reduction, not a guarantee.
  • getTokenLargestAccounts returns token accounts; a big one may be locked LP, a CEX, or a vesting contract — note that rather than calling it a whale.
  • RugCheck score is heuristic and can disagree with the on-chain facts — trust the on-chain authority check over the score when they conflict.
  • This is not financial advice and never a reason to skip user confirmation.

Notes for Hermes

  • Use the terminal tool with python3/curl (not web_extract).
  • HELIUS_API_KEY / SOLANA_RPC_URL flow from ~/.hermes/.env automatically.
  • Read-only: this skill complements — never replaces — the ClawPump MCP (mcp_clawpump_*) trade and launch tools.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.