Bot classes
Taxonomy comes before vendor selection. A product is only fast, cheap or safe relative to what it is trying to do, so this section establishes what each class of Solana automation is for before any comparison of implementations begins.
Most disagreements about which Solana bot is best are disagreements between categories rather than between products. A sniper and a market-activity tool are not competing implementations of one idea: they optimise different quantities, need differently shaped capital, and fail in unrelated ways. Comparing them on a single feature list produces a conclusion that cannot be right.
The pages here do three jobs in order. They define the classes and describe the mechanism each one depends on. They put the classes side by side on the five axes that actually separate them. And they give a framework for judging a specific tool once the class question is settled, which is the only order in which the vendor question is answerable.
What you will not find is a ranked list of products. We compare classes and criteria because a private execution engine cannot be audited from outside, and any score we published would carry more authority than the evidence behind it. The editorial policy explains where that line sits and why it does not move.
Pages in this section
Solana sniper bots explained
The head reference for the class. What sniping means on a chain with no public pending pool, the seven-stage execution pipeline, where detection actually comes from, and a twelve-term glossary.
Trading bot types compared
Sniper, copy, volume and market activity, market making and arbitrage, compared on goal, timing sensitivity, capital shape, venue dependency and dominant failure mode.
Evaluating a trading bot
Nine weighted criteria, a scoring sheet with explicit anchors, the red flags that end an evaluation, and a bounded live-test protocol.
Reading order
If you are new to the subject, read the sniper reference first, because it establishes the vocabulary and the execution sequence that every other page assumes. Then read the comparison to locate your own objective among the five classes. Read the evaluation framework last, at the point where you are about to spend money, since its value is entirely in stopping a purchase you were about to make on incomplete information.
Readers who already know which class they need can skip directly to the evaluation framework, but should still check the overlap discussion in the comparison piece. A surprising share of products sold under one class label are configured versions of another, and the label is the least reliable part of the description.
What belongs in this section, and what does not
Anything that classifies, compares or judges tools belongs here. Anything about the network layer belongs in execution, and anything about security, screening or operational failure belongs in the risk desk. The division is deliberate: the questions that arrive together in a reader's head are answerable in three quite different registers, and mixing them produces pages that are long without being useful.
Start here if you are undecided
The comparison table maps six common objectives onto the class that fits, including the two cases where the honest answer is that no bot class helps.