Detect a memecoin rug

Availability first. The memecoin_packet module ships on an unmerged branch. On a master checkout, every tool in this guide returns module_absent.That is not a reason to skip the guide — the handling of that state is the substance of it.

1. Score the token

get_rug_score(chain="solana", address="...")

2. Branch on `state` before reading anything else

result = get_rug_score(chain="solana", address=token)

if result["state"] != "ok":
    # module_absent | no_snapshot | not_found
    return Unknown(reason=result["reason"])

return Assessment(score=result["data"]["score"])

If your module_absent path is the same code as your ok-with-no-findings path, you have shipped the bug this whole system is designed to prevent.Three of the four states mean “you learned nothing.” Only ok with no findings is weak evidence of safety.

State	Render as
`ok`, no findings	”No risk factors found”
`not_found`	”Not in the scorer’s universe — unchecked”
`no_snapshot`	”Scorer installed but has never run — unchecked”
`module_absent`	”Risk engine unavailable — unchecked”

Never render any of the last three as a green tick.

3. Pull the evidence packet

get_memecoin_packet(chain="solana", address="...")

The packet carries the components behind the composite: authority configuration, liquidity structure, holder concentration, and deployer history — each with its provenance. A composite score without its components cannot be argued with.

4. Check the control plane

Most failures are preceded by an authority change, not a price move: a mint authority reinstated, an upgrade admin rotated, a freeze capability restored.

check_authority_snapshot(chain="solana", subject_id="...")
get_watch_history(subject_id="...")

A subject with no prior snapshot has no changes — which is not the same as a subject that was snapshotted and found stable. See Sentry.

5. Understand why the agent says `no_opinion`

If you query the signal feed for a memecoin on a master deployment, you will get no_opinion with an empty evidence bundle. That is correct behaviour. The evidence collector soft-imports the memecoin collectors, records a named gap when they are absent, and the verifier refuses any verdict the surviving evidence cannot support.

This is worth internalising as a positive signal about the system: on a deployment without the risk engine, it does not generate confident memecoin verdicts backed by nothing. It says no_opinion and records why.

Before deployment

Fail closed

An unreachable risk engine must block, not pass. The default in doubt is “unknown”, never “safe”.

Surface the reason to the end user

“Not checked — risk engine unavailable” is actionable. A silent green tick is not.

Alert on state transitions

A scorer that flips from ok to no_snapshot means your pipeline stopped, and the symptom is silence.

Test the absent path explicitly

Run your integration against a checkout where the module is missing. That configuration is the one that ships to production first.

Scienter publishes impersonal market signals of general and regular circulation. Nothing here is investment, legal, or tax advice, or a recommendation to buy, sell, or hold any asset. Scienter is not a registered investment adviser or broker-dealer. Trading digital assets can result in the total loss of your funds — see Disclaimers for the full text.

Get Started

Concepts

Guides

Detect a memecoin rug

1. Score the token

2. Branch on `state` before reading anything else

3. Pull the evidence packet

4. Check the control plane

5. Understand why the agent says `no_opinion`

Before deployment

​1. Score the token

​2. Branch on state before reading anything else

​3. Pull the evidence packet

​4. Check the control plane

​5. Understand why the agent says no_opinion

​Before deployment

1. Score the token

2. Branch on `state` before reading anything else

3. Pull the evidence packet

4. Check the control plane

5. Understand why the agent says `no_opinion`

Before deployment