# The anti-slop standard

> A written standard for what unedited model output sounds like, and a refusal to ship it.

Canonical: https://clawli.ai/concepts/anti-slop

Readers cannot articulate why a post feels synthetic, but they close the tab anyway. These are the patterns that trigger it.

## Patterns we refuse

- The negation flourish: it is not X, it is Y, used as a substitute for an argument.
- Triads with no content: faster, cheaper, and more scalable.
- Manufactured stakes: this changes everything, the game has changed, here is what nobody tells you.
- Hedged claims that commit to nothing: this can potentially help teams to some degree.
- Em dashes and section emoji used as pacing, in place of sentence structure.
- A closing question that exists only to farm comments.

## What replaces them

A number the author had to earn. A name. A cost that was actually paid. A sentence someone could argue with. The standard is not stylistic preference, it is evidence density: a post that cannot survive removing its adjectives had nothing under them.

## Where it runs

The same standard powers the pre-flight slop check inside the product and the free public checker, which scores a pasted draft and returns the specific lines to fix. No account, nothing stored.

## Related

- [Post formatter and slop checker](https://clawli.ai/tools/post-formatter)
- [Pre-flight](https://clawli.ai/concepts/preflight)
- [The anti-slop standard](https://clawli.ai/blog/anti-slop-standard)