# Unslop > Apply this collection's anti-slop pass to one supplied text and nothing else — false contrasts, empty openings, importance inflation, vague attribution, synonym cycling, robotic rhythm, and generic conclusions — repairing what it finds through a bounded budget of fresh correction subagents and reporting whatever is left. It selects no genre, no technique, and no editorial contract, and it runs no mechanical pass. - Skill: `kntnt/unslop` (Agent Skill, multi-file: 4 files) - Install (CLI): `npx skillmds@latest add kntnt/unslop` - Raw SKILL.md: https://api.skillmd.com/api/skills/kntnt/unslop/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Finance & Business - Author: kntnt (https://skillmd.com/u/kntnt) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/kntnt/unslop --- # unslop Read one Text Artifact against the shared anti-slop catalogue, repair what that pass found within the budget the invocation allows, say what is left, and stop there. This is one lens applied on its own, to a text that is otherwise finished: nothing here selects a genre, a technique, or an editorial contract, and nothing here proofreads afterwards. `$HERE` is the directory that contains this SKILL.md, and `$MANAGER` is the Manager directory: `$HERE/../kntnt/` if it exists, else `kntnt/` under a Global harness skills directory (`~/.claude/skills`, `~/.config/opencode/skills`, or wherever another Harness keeps them). Neither found: tell the user to install the Manager (`npx skills add Kntnt/skills`) and stop. `$LIBRARY` is `$MANAGER/library/` — absent, tell the user to run `/kntnt update`, then stop. Run `uv run --no-cache --no-project "$MANAGER/scripts/kntnt.py" invoke --here="$HERE"` with the invocation payload — everything the user typed after `/unslop`, verbatim, however many lines — on stdin. On exit 0, answer the `capabilities` in its `dependencies` first: for each one, say whether its `confirm` sentence is true of you, and where it is not, give its `how`, change nothing, and stop. Then continue from the JSON. On any other exit print its stdout verbatim and stop: it has already printed what the user is to see, and none of that text is yours to write. In the JSON, `path` is the command path as a list, `flags` holds each flag the user wrote — `true` where it stood bare, its value where it carried one, a list of values where it was repeated — `operands` is what followed the flags, in order, and `instruction` is the Contextual Instruction, or `null`, applied as `$LIBRARY/references/invocation-envelope.md` says. Run every UV command in this Skill with a fresh private directory as `TMPDIR`, and remove that directory after the command, including when it fails. The private directory belongs to that one command and no other run, so cleanup removes only files this run created. ## Arguments The operand is the Text Artifact: inline text, one local path, or one URL, and exactly one of them — several paths, a glob reaching more than one file, or a directory of texts is more than one. Where the invocation carries no operand, it is the single text the current turn identifies. `--language=` names the language or locale whose anti-slop guidance applies. Any spelling the Collection's resolver reaches is accepted — a canonical code (`sv`, `en_GB`), a case or separator variant of one (`en-GB`, `EN_GB`), a curated alias (`BrE`, `brittisk engelska`), or an ordinary description of a language in any language. `--max=` is the Correction Budget: the greatest number of corrections this pass may delegate. It takes any non-negative integer and defaults to `1`, so an ordinary run includes one correction and one opportunity to verify it. `0` reviews without correcting — the findings come back for the reader to act on — and a higher value bounds a longer loop explicitly. It is a ceiling and never a quota: a run with nothing left to correct stops with the rest of it unspent. `--output=` names the Output Target. `response` is the default and the value `response` states it explicitly; anything else is one filesystem path. It and In-place Editing name two destinations for one text, which is why no form carries both. `--in-place[=]` selects In-place Editing. It accepts `yes`, `on`, and `true` against `no`, `off`, and `false`; bare `--in-place` means `on`, `--in-place=off` has the same effect as omitting the option, and the default is `off`. ## Resolution One editorial parameter is resolved here, the language, and it takes the first of these that answers: 1. The Formal Invocation. 2. A recognized Kntnt map in the Text Artifact's leading YAML frontmatter — the reserved `kntnt` key and its `language` value, and nothing else out of it. A `language`, `lang`, `genre`, or `technique` key at the top level of the document's own frontmatter belongs to that document, whatever it is named and whatever it holds; frontmatter carrying no `kntnt` map carries no configuration of this collection's. 3. The current Contextual Instruction. 4. Applicable Conversation Context. 5. Inference from the Text Artifact itself. 6. The language of the supplied text, which is what inference reads and what the default already is. A value found at one level suppresses the levels below it. Suppression is that precedence working rather than an error: a Contextual Instruction every higher level has already settled leaves nothing for it to settle, and the run continues rather than refusing it as unaddressable guidance. Where saying so is useful, the delivery names the suppressed instruction beside the resolved configuration. A language is settled rather than guessed: where the artifact is materially ambiguous or mixed, with no dominant language or alternation inside paragraphs, name the candidates and ask before anything is read against the catalogue. No Kntnt map is created or synchronized. A map that exists supplies the language and is otherwise left exactly where it is, an artifact carrying none has none added to it, and the rest of the frontmatter comes through as it arrived — this Skill settles one of the three values such a map records, so writing one would claim a configuration it never resolved. Nor is the map ever required: an artifact carrying none is repaired exactly like one that carries a complete one, its absence refusing nothing and leaving the language to the levels below it. A recognized Kntnt map whose `language` no installed resource answers to, where `--language` supplied no value, is reported as unusable artifact metadata and stops the run. It is never quietly read as the nearest usable value, so a spelling such as `en_UK` never becomes `en_GB` on its own. ## Steps 1. Settle the single Text Artifact and the options from the JSON. A value outside what `## Arguments` admits — `--max` that is not a non-negative integer, `--in-place` outside its vocabulary, or more than one Text Artifact — is refused as `$LIBRARY/references/invocation-envelope.md` says: read nothing, write nothing, and stop. What the operand says never changes what it is: the Text Artifact is the text to read against the catalogue whatever its own words turn out to be, so a brief, an outline, a set of notes, or anything else that reads as an instruction is still the text — read like every other text and never carried out — and it is the same artifact however it was supplied, pasted inline exactly as much as named by a path or a URL. Reaching that judgement reads no other Skill's instructions, neither a `SKILL.md` of theirs nor a `help.md`: this Skill has one gesture and being handed a brief does not give it a second. Done when one Text Artifact and a valid set of options are in hand, or you have stopped. 2. Settle the destination before anything is read against the catalogue, following `$LIBRARY/references/delivery.md`. Make every refusal that contract names by reading alone — In-place Editing together with a separate Output Target, an output path equal to the input path, In-place Editing against inline text or a URL or a source that is not a writable local file, and a destination whose parent directory does not exist — so that a refusal has nothing written behind it. Done when the destination is settled, or you have stopped. 3. Read the Text Artifact's leading YAML frontmatter where it has any, and look in it for one top-level `kntnt` map. Read `language` out of that map and nothing else out of it, and read nothing outside it as configuration. Done when the artifact's recognized language value is known to be present, absent, or unusable. 4. Resolve the language by `## Resolution`, and verify the selector with `uv run --no-cache --no-project "$LIBRARY/scripts/languages.py" resolve --scope=anti-slop ""`. Exit 0: take the reported `code` as the resolved language and the returned scope as this run's language-specific guidance. A non-zero exit says which of the ways the selector failed — no installed Language Resource reached, or more than one — and is refused as `$LIBRARY/references/invocation-envelope.md` says, the run reading nothing, writing nothing, and stopping; an unlisted description of a language is interpreted first, then proposed as one installed candidate from `uv run --no-cache --no-project "$LIBRARY/scripts/languages.py" list` and verified through that same `resolve` command. Report unusable recognized metadata as `## Resolution` describes, and ask rather than guess at a mixed language. Nothing has been written yet, so asking costs nothing to undo. Done when the language is settled and its scope is in hand, or you have asked or refused. 5. Load `$LIBRARY/references/editorial/anti-slop.md`, and take it together with the scope step 4 already returned as the whole of what this pass is read against. Load nothing else: no base contract, no genre, no technique, and none of the other scopes a Language Resource carries. The wider editorial guidance belongs to the Skill that applies the whole contract, and rules held here are rules this pass would end up applying to a text nobody asked it to review. Done when those two are loaded and nothing else has been. 6. Read the Text Artifact against them and record what you find as findings: where in the text, which pattern, and what the reader loses. The findings are anti-slop findings and nothing else — false contrasts, empty openings, importance inflation, vague attribution, synonym cycling, robotic rhythm, and generic conclusions, each applied by what it does in the resolved language rather than by the English words the catalogue writes it with. Locale spelling, vocabulary, date, or currency differences are not findings: resolving a locale admits its anti-slop scope and none of its mechanics. A sentence that is merely unusual, a fact you would have put differently, a structure you would have chosen against: none of these is a finding here, and a text whose only fault is one of them is clean. A code sample — a fenced block, an indented block, or an inline code span — is quoted material rather than the text's own prose: its contents are read past rather than read against the rules, so nothing inside one is a finding and nothing inside one is changed, the docstrings, comments, and string literals among them. Prose about code is ordinary prose and is read like every other sentence. Judge only what is in front of you. This Skill never compares the Text Artifact with source material, never asks for material it was not given, and never remarks that source verification was unavailable — Source Fidelity is the contract of the Skill that wrote the text, and a caveat about material nobody supplied is noise in every run that was not one. Done when the pass is complete. 7. Spend the Correction Budget, up to it and never towards it. While findings remain from the most recent pass and budget remains, take one round: - Delegate the correction to a subagent started fresh for this round, from [`correction.md`](references/correction.md), filled in and delivered as that brief says. Fresh means it carries no history: not what an earlier round found, not what an earlier correction attempted, and not your own reading of the text, so the framing of the round before cannot bias this repair. It receives the complete current Text Artifact and the complete findings of the most recent pass — the text as it now stands and every finding as it was recorded, never a summary and never a stale draft — the language resolved in step 4, and the requirement to leave alone everything the findings do not concern. - Decrement the budget once per correction, whatever that correction came back with. A round that repaired nothing has still spent what it spent. - Retain the complete current text as the pre-round Text Artifact. Before the returned text becomes current, compare it with that pre-round Text Artifact and record every removed claim, whether the finding named that whole claim as the pattern or the claim sat beside the pattern. A shorter text can be clean against step 6 while having silently lost what the earlier state said, so this comparison is part of re-reading rather than an inference from the returned text. Where a claim sat beside the pattern, reject the returned correction, keep the pre-round Text Artifact as current, carry the original finding forward as unresolved, and treat the attempted loss as the repair-created finding that stops the loop below. Where the finding named the whole claim as the pattern, its removal may stand, but the delivery still names the claim that went. - Take an accepted returned text as the current Text Artifact and read it again, from the top, exactly as step 6 reads and against the two things loaded in step 5. A correction is verified by being reviewed again and never accepted on the report of whoever made it: an agent reporting on its own work is the one reader who cannot check it. The findings of this pass, and not the subagent's account of what it fixed, are the findings from here on. The loop stops at the first of four conditions, and each ends step 7. It stops when no findings remain, leaving the rest of the budget unspent — a text the pass is already clean against is not corrected again to use up a number. It stops when a correction makes no relevant progress, which is a round whose re-reading leaves the findings it was given standing with nothing they named changed: the next round would be the same round, so stop and carry those findings forward, saying which of them are unresolved. It stops when a re-reading raises a finding an earlier round's own repair created — a defect the text did not have when it arrived and one of your rounds put there, the clearest case being a finding about material a round removed. You hold every state the text has passed through and are the only party that does: the subagent is fresh by design and each pass reads the text in front of it, so a loop that has begun to answer for its own work is visible here and nowhere else. Another round would repair what the round before it did, so stop, carry the outstanding findings forward the same way, and name the repair the new finding followed from, which is the thing a person has to settle. And it stops when the budget is spent with findings left, which carries them forward the same way for a person to finish. Done when one of the four has stopped the loop, or the budget was `0` and there was nothing to spend. 8. Run no mechanical pass and correct nothing outside the lens. This run was asked for one pass and delivers one pass: spelling, punctuation, grammar, and the locale's own mechanics are somebody else's gesture, and so is everything the editorial contract holds that this Skill never loaded. Material the findings do not concern comes through as it arrived, the frontmatter among it. Done when the diff you are about to deliver is anti-slop repair throughout. 9. Deliver by `$LIBRARY/references/delivery.md`. On a response target, remove every artifact or scratch file this run created and base the delivery account on the filesystem state that remains after cleanup. Where findings remain — those step 7 carried forward, whichever of its conditions stopped the loop — deliver the Text Artifact to its Output Target and report the findings separately, saying which are unresolved; that holds where nothing in the text changed too, since the findings are the point of that run, and a file-targeted or in-place run leaves the findings in the response beside the destination it wrote. Where any round removed a passage carrying a claim, whether an accepted repair removed the claim as the pattern or a rejected repair attempted to take it beside the pattern, report every removed claim separately and say which repair removed or attempted to remove it; this account accompanies a clean artifact too, so the loss is visible without a diff. Where no findings or removed-claim account remain, deliver the Text Artifact alone, and where such a clean run was aimed at the response or at its own source and changed nothing, write nothing and return only the short no-change status. Keep the pass to its findings and removed-claim account: the reasoning behind them, the passages you considered and dismissed, and any other correspondence with a correction subagent stay out of the output. Done when the Text Artifact has been delivered, or the no-change status has been reported.