agentleFS
leonardourcileonardourci/pnyx/.agents/skills/deslop/SKILL.md

deslop

Skill4 starsChanged 3 days agoApache-2.0View on GitHub

Remove AI-style wording, em dashes and needless complexity from Pnyx text, including code comments, docs, UI copy, PR descriptions and commit messages.

129 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use control-cli, openseo-review-web-content and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ElyzeSolutionsSkill

    control-cli

    Build or adapt a local harness to drive, inspect, and profile an interactive CLI or TUI without external services. Use for CLI UX checks, startup regressions, memory leaks, hangs, prompt flows, or terminal demos.

    Also in 6 projects that use this

  • JwuthriSkill

    openseo-review-web-content

    Write and review content for the OpenSEO website (web/) — blog posts, guides, feature pages, FAQs. Distills the philosophy for on-brand, useful, accurate content. Use whenever adding or editing user-facing prose in web/content or web/src.

    Also in 4 projects that use this

  • automagik-devSkill

    genie-hacks

    Browse, search, and contribute community hacks — real-world patterns for provider switching, teams, skills, hooks, cost optimization, and more.

    Also in 3 projects that use this

  • automagik-devSkill

    workfly

    Discover a procedure and build its saved workflow — dynamic discovery, drafted script, adversarial verification, landed in the catalog.

    Also in 3 projects that use this

  • JwuthriSkill

    simple-issue-description

    Turn a rough bug report, feature request, support note, or pull request into a short, plain-language issue focused on the problem and desired behavior. Use when a contributor asks to simplify an issue, explain what a PR is for, create the corresponding issue for a PR, remove implementation detail from a report, or invokes /simple-issue-description.

    Also in 4 projects that use this

  • learnerLjSkill

    verify-this

    Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

    Also in 6 projects that use this

  • +2 more from Jwuthri
  • +3 more from automagik-dev

The file

Files
SKILL.md

Deslop

Apply this checklist to the changed text, whatever its format. Preserve facts, contracts, necessary caveats and the writer's intended meaning. Leave licenses, attributed quotations and external protocol data intact.

  • Read as someone who did not see the conversation. Lead with the concrete problem, action or result. Remove drafting history, assistant chatter, self-praise and summaries that repeat the preceding paragraph.
  • Replace vague claims and AI filler (“delve”, “leverage”, “seamless”, “robust”, “it is worth noting”, “this PR aims to”) with plain words and specific facts. Do not add claims the evidence does not support.
  • Remove em dashes from authored text. Use a period, comma, colon or parentheses that fits the sentence. Keep punctuation natural rather than mechanically replacing every dash with a hyphen.
  • Cut contrastive slogans, invented labels, unnecessary headings and repeated explanations. Keep lists when they help someone follow steps or compare parallel facts.
  • Keep each prose paragraph on one source line. Use line breaks for lists, code and separate paragraphs.
  • In code comments, retain non-obvious reasons, invariants, compatibility constraints and useful public contracts. Remove narration, step numbers, signature paraphrases, stale claims and comments that repeat test names. Prefer clearer names when they carry the meaning.
  • When code is in scope, remove dead branches, unused scaffolding and one-use wrappers that obscure the operation. Avoid introducing abstractions or dependencies just for cleanup. Keep useful collection methods, concurrent work and declarative rendering.
  • Preserve error visibility, ordering, cancellation, retries, cleanup and permissions. A wording pass must not quietly change behavior or hide a failure.
  • Check Pnyx vocabulary and update stale usage guidance in the root README. Use Hydra, Oracle and inscribe; keep other terms plain.
  • Read the final diff, including the PR description and proposed commit messages. Run git diff --check; for executable changes run focused checks and pnpm verify as allowed. A prose-only cleanup needs no new behavior tests.

Code cleanups

A cleanup removes complexity the job didn't ask for without changing what the code does. follow-conventions holds the target style, and the Simpler part of review-code lists what a review flags. Clear beats short: the test is whether a new reader follows the code on first read, not the line count. Three plain named steps beat one chained line.

  • Unpack cryptic shorthand in new and changed code into plain named steps: conditional spreads (...(x ? { a } : {})), logical assignment picking a value (??=, ||=, &&=), || or ?? defaults whose case the names don't make obvious, a bare void on a promise, .then chains and for (;;) or while (true) waits. Keep every arm's outcome: a || 'none' that fires on an empty list becomes a named empty-list check. Existing uses elsewhere are not rewritten.
  • Leave code that says what it wants alone. map and filter over a list, Promise.all for independent work and a JSX condition that picks what shows are not clutter; don't rewrite them into step-by-step loops or one-at-a-time awaits. Bad: const titles = []; for (const card of cards) { if (card.open) { titles.push(card.title) } }. Good: const titles = cards.filter(card => card.open).map(card => card.title).
  • Use static imports. A dynamic import() requires a comment explaining its real need, such as a module that must load lazily.
  • Reuse what exists: replace a copy of an existing helper, component or platform feature with it, and name what keeps working.
  • Remove indirection; don't add it. No new abstraction, helper layer or dependency to make a pattern look cleaner. A named helper still belongs where it names a distinct read, decision or write, or one decision used in several places. A one-off fallback stays a local try/catch at that spot.
  • Review file splits using follow-conventions. Recommend changing a split that adds indirection without a clear job or scatters closely related code.
  • Keep every path's outcome. When a .catch(() => fallback) becomes try/catch, the fallback value and the errors that reach it stay the same. Serialized work stays serialized, independent work stays concurrent, and retries, cleanup in finally and the message the user sees stay as they were.
  • Keep the diff small and mechanical, one logical change per commit, so a reviewer can compare old and new arm by arm. Update or delete guidance the cleanup made stale in the same change.
  • Scan what you added for new shorthand: git diff origin/main -U0 -- '*.ts' '*.tsx' | grep -E '^\+.*(\.then\(|\bvoid [a-zA-Z(]|\?\?=|\|\|=|&&=|\.\.\.\(|for \(;;\)|while \(true\))'. Name any you keep on purpose in the PR, with its reason.

Record meaningful fixes and any intentional exception. Once complete, tick Ran deslop in the PR and note what you changed (or say no findings).

For an agent: edit the assigned text yourself, keep the author's meaning, and report any validation gap honestly. This checklist grants no extra authority to commit, publish or send text.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.