{"id":50384887,"url":"https://github.com/0plus1/non-fiction-workbench","last_synced_at":"2026-05-30T14:30:27.071Z","repository":{"id":352156946,"uuid":"1214069070","full_name":"0plus1/non-fiction-workbench","owner":"0plus1","description":null,"archived":false,"fork":false,"pushed_at":"2026-04-18T05:25:11.000Z","size":14,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-18T07:25:45.726Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/0plus1.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-18T04:46:26.000Z","updated_at":"2026-04-18T05:25:14.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/0plus1/non-fiction-workbench","commit_stats":null,"previous_names":["0plus1/non-fiction-workbench"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/0plus1/non-fiction-workbench","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0plus1%2Fnon-fiction-workbench","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0plus1%2Fnon-fiction-workbench/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0plus1%2Fnon-fiction-workbench/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0plus1%2Fnon-fiction-workbench/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/0plus1","download_url":"https://codeload.github.com/0plus1/non-fiction-workbench/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/0plus1%2Fnon-fiction-workbench/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33696681,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-30T02:00:06.278Z","response_time":92,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-05-30T14:30:25.305Z","updated_at":"2026-05-30T14:30:27.067Z","avatar_url":"https://github.com/0plus1.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Non-Fiction Workbench\n\nLLM agent skills for long-form non-fiction writing and revision.\n\n- `/non-fiction-workbench` for drafting, rewriting, and critiquing non-fiction against a brief, source set, and house style.\n- `/anti-slop` for detecting and removing common LLM prose tells without changing factual meaning.\n\n## Why This Exists\n\nAI-assisted non-fiction usually fails in predictable ways:\n\n- claims get stronger than the evidence\n- structure drifts away from the brief\n- paragraphs bloat with transitions, summary, and throat-clearing\n- rewrites flatten voice and introduce generic \"smart\" phrasing\n\nThis repo gives agents a tighter workflow for essays, articles, reported features, newsletters, and book chapters. It does not replace authorship or reporting. It gives the model a disciplined way to read the draft, inspect the brief and source material, and revise without hallucinating or padding.\n\nThe anti-slop skill is informed by common LLM prose-tell catalogs, including the public taxonomy at [tropes.fyi](https://tropes.fyi/tropes-md), then distilled into rewrite-oriented rules for actual editorial use. It also includes a stricter parity checklist for detector-style passes, based on the rule inventory used in [Slop Cop](https://github.com/awnist/slop-cop).\n\n## What's Included\n\n```text\nnon-fiction-workbench/\n├── README.md\n├── LICENSE\n└── skills/\n    ├── non-fiction-workbench/\n    │   ├── SKILL.md\n    │   ├── agents/\n    │   │   └── openai.yaml\n    │   └── references/\n    │       └── project-shape.md\n    └── anti-slop/\n        ├── SKILL.md\n        ├── agents/\n        │   └── openai.yaml\n        └── references/\n            └── slop-patterns.md\n```\n\n## Install\n\nInstall the core writing skill:\n\n```bash\nnpx skills add https://github.com/0plus1/non-fiction-workbench --skill non-fiction-workbench\n```\n\nInstall the anti-slop skill:\n\n```bash\nnpx skills add https://github.com/0plus1/non-fiction-workbench --skill anti-slop\n```\n\n## Usage\n\nInvoke the core skill manually:\n\n```text\n/non-fiction-workbench write drafts/article.md\n/non-fiction-workbench rewrite drafts/article.md\n/non-fiction-workbench critique drafts/article.md\n/non-fiction-workbench global-critique drafts/\n/non-fiction-workbench global-rewrite drafts/\n```\n\nInvoke the anti-slop skill manually:\n\n```text\n/anti-slop scan drafts/article.md\n/anti-slop revise drafts/article.md\n/anti-slop global-scan drafts/\n/anti-slop global-revise drafts/\n```\n\n## Modes\n\n`non-fiction-workbench`\n\n- `write`: draft or expand a non-fiction piece from the brief, outline, and source material\n- `rewrite`: improve an existing draft while preserving thesis, facts, and structure\n- `critique`: return editorial feedback on argument, evidence, structure, and prose\n- `global-critique`: critique a whole draft directory as one manuscript\n- `global-rewrite`: rewrite a whole draft directory while preserving argument and evidence\n\n`anti-slop`\n\n- `scan`: flag clear AI-style rhetoric and document-level repetition\n- `revise`: rewrite to remove AI tells while preserving factual content and argument\n- `global-scan`: scan a full draft directory for repeated AI-style patterns across the manuscript\n- `global-revise`: remove slop across a full draft directory while preserving meaning\n\n## Expected Project Shape\n\nThe skills are opinionated about project structure. They work best when your project looks roughly like this:\n\n```text\nyour-book-or-essay/\n├── brief/\n│   ├── thesis.md\n│   ├── audience.md\n│   ├── constraints.md\n│   └── outline.md\n├── research/\n│   ├── source-index.md\n│   ├── notes/\n│   └── excerpts/\n├── style/\n│   ├── voice.md\n│   └── house-style.md\n└── drafts/\n    ├── chapter-1.md\n    ├── essay-1.md\n    └── article-x.md\n```\n\nThe core skill reads the target draft first, then consults the relevant brief, style, and source files. For manuscript-level work, it can read an entire `drafts/` directory in order. The anti-slop skill reads the target draft and applies a conservative rule set for common LLM prose patterns, including repeated rhetorical scaffolds that become obvious only at manuscript scale.\n\n## How They Behave\n\nBoth skills are deliberately manual. They use `disable-model-invocation: true`.\n\nThey tell agents to:\n\n- read the target draft first\n- inspect the relevant brief, style, and source files before changing claims or structure\n- preserve the author's thesis, evidence, and level of certainty\n- avoid inventing facts, sources, examples, chronology, or causal claims\n- cut generic LLM phrasing, padding, and rhetorical scaffolding\n- keep the human in charge of judgment, reporting, and final wording\n\nFor `write` and `rewrite`, the core skill is intended to revise the target draft directly when appropriate. For `critique` and `scan`, it should usually return notes unless you explicitly ask it to write those notes into project files.\n\n## Philosophy\n\nThis repo is not an autopilot content generator.\n\nThe author still owns:\n\n- thesis\n- reporting\n- structure\n- taste\n- final language\n\nThe agents' role here is narrower:\n\n- reduce factual drift\n- keep the brief and source material in view\n- make revision more disciplined\n- remove generic AI prose habits before publication\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0plus1%2Fnon-fiction-workbench","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F0plus1%2Fnon-fiction-workbench","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F0plus1%2Fnon-fiction-workbench/lists"}