05 · Memecoins
New-pair rug screen
A reject-first screen for a fresh pair: mint and freeze authority, liquidity control, taxes, holder concentration, and unverified code.
- Level
- Starter
- Cadence
- Once per pair, before any size you cannot lose
- Read
- 9 min
- Updated
Definition
A new-pair rug screen is a sequence of checks that try to kill a fresh pair: mint authority, freeze authority, proxy admin, unverified code, liquidity the deployer can pull, transfer taxes, and holder concentration. Surviving the screen is not a buy. It only earns a second question about whether a product exists.
Who it is for
Traders who look at new pairs and want a mechanical reject list before anyone says the word narrative.
Job to be done
Run a reject-first screen on one pair and stop at the first hard fail, with sources you can reopen.
Human gate. A hard fail is a pass on the trade. You do not ask the agent for a reason to ignore a fail. Only a clean screen may be handed to the product check.
Inputs and outputs
Inputs
- Chain, pair, and token mint you copied from the venue you would trade
- The deployer or pool address if you already have it
- Your hard-fail list, written before the screen
- Whether you will allow a tiny test sell, sized as money you can lose
Outputs
- Pass or hard-fail on each check, with the source
- The first hard fail called out at the top
- Unknowns listed instead of assumed safe
- No price target and no "early" language
Step-by-step workflow
- 01Copy addresses from the venueChain, pair, mint. Do not let a social post supply the mint. One wrong mint makes the rest of the screen fiction.
- 02Check control of supplyMint authority, freeze authority, and whether the token is a proxy with an admin. Unverified code is a fail or an unknown you refuse to treat as a pass.
- 03Check control of liquidityWho holds the LP tokens, and whether a lock is visible and irreversible. "Trust me" from the deployer is not a lock.
- 04Check whether a sell worksRead displayed taxes. If you use a simulator or a tiny buy-and-sell, size it as money you are willing to lose, and record the result.
- 05Check holdersConcentration outside the pool. A pile of fresh wallets is a question, not a community. Do not invent identities for them.
- 06Human gateIf any hard fail fired, stop. If none fired, the only next playbook is the product screen, still not an order.
Suggested tool types
These are categories of tools, not product picks. Crypto Agent Desk does not certify vendors and does not run an affiliate program.
| Type | Why it is here |
|---|---|
| Block explorer | Mint authority, freeze, verification, proxy, and holder list. |
| Pool or DEX page | LP token holder, lock claims, and tax fields the venue displays. |
| Contract simulator or a tiny test trade | A sell that fails is information. Size the test as a total loss. |
| Chat model | Holds the reject order so a story cannot jump the queue. |
Risks and failure modes
Renounce theater
Mint authority can look renounced while a proxy or a second role can still change the token. If you cannot read that, the check is unknown, not pass.
Locked liquidity that is not locked
A locker the owner can unlock, or LP sent to a wallet that is mislabeled "burn", is not a lock. Record the LP holder exactly.
Screen as a blessing
A pair that passes can still go to zero for ordinary reasons. The screen only removes some mechanical traps.
Operator notes
New pairs are a reject problem. The screen is ordered so the story does not get a vote until the controls are checked.
Hard fails to set before you start
Unverified code you cannot read. Active mint or freeze you did not accept in writing. LP held by the deployer with no irreversible lock. A test sell that fails. A tax that can be changed by an admin you do not control. Any one of these ends the screen.
Prompt you can paste
Rug-screen this pair. Chain: CHAIN. Pair: PAIR. Mint: MINT.
Hard fails I already accept: FAIL_LIST.
Use the notes I paste from the explorer and the pool page: NOTES.
Order: supply control, liquidity control, tax or sell behavior, holder concentration.
Put the first hard fail at the top. Unknown is not a pass.
Do not discuss narrative, price targets, or why it could "still work".
If it survives, say only: eligible for a product screen, not a buy.Done when
The top line is a hard fail with a source, or a short list of passes and unknowns that you are willing to take into the product screen.
FAQ
Does a passed screen mean the token is safe?
No. It means a few ways of getting trapped or diluted did not show up in the checks you ran. Price, liquidity, and honesty of the social layer are separate.
What if the contract is unverified?
Treat it as a fail under this playbook unless you can read the bytecode yourself, which this page does not pretend to teach. Unknown is not a pass.
Should I skip the screen if the narrative is strong?
No. Narrative is the next page, and only if this one does not fail. A strong story is not evidence about mint authority.
Published . Updated . This is an operating playbook, not a recommendation to buy, sell, or hold any asset.