Anti slop slop canon
Skill codeSTACKr/anti-slop-slop-canon/skills/anti-slop-slop-canon
Write, compose, rewrite, edit, or audit everyday text such as notes, updates, posts, emails, docs, scripts, narration, and voice-agent replies. Applies your voice without AI-writing habits. Also covers voice onboarding or setup, profile inspection or refresh, and realtime style-prompt requests.From its SKILL.md
npx -y skills add codeSTACKr/anti-slop-slop-canon --skill anti-slop-slop-canonAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 24 days oldThe repository was created 24 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.0 KB, 376 tokens by cl100k_base, as published. Nobody here has run it
Anti-Slop Slop Canon
Use default content version 0.1.0.
Resolve scope and one bundle
- In a project copy, inspect only
<project-root>/.anti-slop-slop-canon/; never inspect or fall back to global state. Otherwise inspect only~/.config/anti-slop-slop-canon/. - Validate
voice-profile.mdwith references/profile-schema.md. Require schema1.0.0, active scope, every required section, and actionable rules. Load it alone; otherwise load references/defaults.md alone. Report invalidity after the task and offer a previewed repair. - When profile and
settings.mdare absent, gate the first natural-language task: read references/onboarding.md and offer personalize, defaults, or defer before writing. Also read it for explicit onboarding, setup, or voice learning; otherwise do not. - With a profile, compare its loaded
defaults_versionto0.1.0without reading defaults. Finish the task with that profile before any notice or lifecycle action.
Protect and route
- Select spoken only when text is clearly meant to be heard. Otherwise use written without asking.
- Preserve meaning, facts, uncertainty, intent, exact quotations, code, structured data, and legally fixed wording. Reject detector-evasion framing and offer clarity or voice editing.
- Read references/operations.md. Route explicit operations and lifecycle choices; infer compose for creation and rewrite for revision.
- Apply one bundle and mode. Silently check meaning, exemptions, mode, and rules, then fix only mutable prose.
- Return clean compose or rewrite output. Add reasoning or a diff only when requested. Audit without editing.
What ships with it: 7 files
24.8 KB alongside SKILL.md
agents/
- openai.yaml229 B
assets/
references/
- defaults.md5.4 KB
- onboarding.md6.3 KB
- operations.md8.1 KB
- profile-schema.md2.4 KB