[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79S657JZHVCQRJA2X4CT6WS":3,"skill-same:01M79S657JZHVCQRJA2X4CT6WS":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M79S657JZHVCQRJA2X4CT6WS","review-writing-slop",null,"Review or revise prose with the SlopScoreTool (sst) analyzer while preserving purpose and voice. Use when drafting, editing, comparing revisions, investigating repetitive or canned AI-associated language, or when a user explicitly asks to reduce AI slop. Do not use it to determine whether AI wrote a text.","lemon07r\u002FSlopScoreTool","lemon07r","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F12001338?v=4","01M79S5XPN8PC6JRRC6WHNAHRC","Lamim","01M79S5XQ67NAYY3K4DQBE0WFM",2,"MIT","https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool\u002Fblob\u002Fmain\u002Fcmd\u002Fsst\u002Fskill\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: review-writing-slop\ndescription: Review or revise prose with the SlopScoreTool (sst) analyzer while preserving purpose and voice. Use when drafting, editing, comparing revisions, investigating repetitive or canned AI-associated language, or when a user explicitly asks to reduce AI slop. Do not use it to determine whether AI wrote a text.\n---\n\n# Review Writing Slop\n\nUse the analyzer as editorial evidence, not as an authority or optimization target.\n\n## Setup\n\nThe `sst` binary must be on `PATH`. If it is missing, install it from the\nproject releases page or with\n`go install github.com\u002Flemon07r\u002FSlopScoreTool\u002Fcmd\u002Fsst@latest`. Inside the\nsource repository, build with `make build` if the binary is absent. If the\ninstalled version exposes a different interface, inspect `sst --help` rather\nthan guessing.\n\n## Analyze\n\n1. Establish the draft's purpose, audience, genre, voice, and hard constraints.\n2. Analyze a substantive draft. Do not compose from a prohibited-word list.\n3. Request bounded, agent-optimized output:\n\n   ```sh\n   sst --format agent --max-findings 20 [FILE...]\n   ```\n\n   With no files, pass the draft on stdin. Use `--format brief` instead when a\n   structured pipeline consumes the JSON; it is the same evidence with\n   provenance, calibration detail, and excerpts stripped. Reserve\n   `--format json` for the full-fidelity contract.\n\n4. Read warnings first: short samples, few words, invalid UTF-8, and truncated\n   findings all limit confidence. When output reports truncated findings,\n   rerun with `--full` before concluding a pattern is absent.\n5. Inspect the strongest repeated findings in their surrounding prose. The\n   report gives byte offsets (`@start:end`) into the original text; re-read\n   the source around them rather than judging phrases in isolation. Treat\n   isolated low-confidence matches lightly. Use `--exclude-term` only for\n   vocabulary that is necessary to the subject (names, domain terms), never\n   to make findings disappear.\n\nFor a before\u002Fafter question, prefer `sst --compare BEFORE AFTER` over two\nseparate runs; it reports score, component, and finding deltas without\ndeclaring a winner.\n\n## Judge\n\n- Ask whether each finding reveals repetition, empty emphasis, vague\n  abstraction, redundant explanation, canned structure, or another weakness a\n  reader would notice without the metric.\n- Keep language that is precise, intentional, genre-appropriate, rhythmic,\n  quoted, or necessary to the subject.\n- Give repeated patterns and concentrated habits more weight than a single\n  flagged phrase. In multi-file runs, shared findings mark author-level\n  habits, not topic vocabulary.\n- Never infer authorship, competence, honesty, or overall writing quality\n  from a score.\n\n## Revise\n\nRevise only findings that independently improve clarity, specificity,\neconomy, rhythm, or voice. Prefer fixing the underlying sentence or paragraph\nover mechanically swapping a flagged word. Preserve useful texture and\nunevenness.\n\nFor a meaningful revision, optionally compare drafts directly:\n\n```sh\nsst --format agent --compare BEFORE.txt AFTER.txt\n```\n\nRead the changed findings in context rather than the score delta. Stop when\nthe prose is better. Do not chase a target score, zero findings, or endless\npasses.\n\nReturn the improved text when revision was requested. Summarize only the\neditorial choices that help the user; mention important findings deliberately\nretained when that explains the result.\n","# Review Writing Slop\n\nUse the analyzer as editorial evidence, not as an authority or optimization target.\n\n## Setup\n\nThe `sst` binary must be on `PATH`. If it is missing, install it from the\nproject releases page or with\n`go install github.com\u002Flemon07r\u002FSlopScoreTool\u002Fcmd\u002Fsst@latest`. Inside the\nsource repository, build with `make build` if the binary is absent. If the\ninstalled version exposes a different interface, inspect `sst --help` rather\nthan guessing.\n\n## Analyze\n\n1. Establish the draft's purpose, audience, genre, voice, and hard constraints.\n2. Analyze a substantive draft. Do not compose from a prohibited-word list.\n3. Request bounded, agent-optimized output:\n\n   ```sh\n   sst --format agent --max-findings 20 [FILE...]\n   ```\n\n   With no files, pass the draft on stdin. Use `--format brief` instead when a\n   structured pipeline consumes the JSON; it is the same evidence with\n   provenance, calibration detail, and excerpts stripped. Reserve\n   `--format json` for the full-fidelity contract.\n\n4. Read warnings first: short samples, few words, invalid UTF-8, and truncated\n   findings all limit confidence. When output reports truncated findings,\n   rerun with `--full` before concluding a pattern is absent.\n5. Inspect the strongest repeated findings in their surrounding prose. The\n   report gives byte offsets (`@start:end`) into the original text; re-read\n   the source around them rather than judging phrases in isolation. Treat\n   isolated low-confidence matches lightly. Use `--exclude-term` only for\n   vocabulary that is necessary to the subject (names, domain terms), never\n   to make findings disappear.\n\nFor a before\u002Fafter question, prefer `sst --compare BEFORE AFTER` over two\nseparate runs; it reports score, component, and finding deltas without\ndeclaring a winner.\n\n## Judge\n\n- Ask whether each finding reveals repetition, empty emphasis, vague\n  abstraction, redundant explanation, canned structure, or another weakness a\n  reader would notice without the metric.\n- Keep language that is precise, intentional, genre-appropriate, rhythmic,\n  quoted, or necessary to the subject.\n- Give repeated patterns and concentrated habits more weight than a single\n  flagged phrase. In multi-file runs, shared findings mark author-level\n  habits, not topic vocabulary.\n- Never infer authorship, competence, honesty, or overall writing quality\n  from a score.\n\n## Revise\n\nRevise only findings that independently improve clarity, specificity,\neconomy, rhythm, or voice. Prefer fixing the underlying sentence or paragraph\nover mechanically swapping a flagged word. Preserve useful texture and\nunevenness.\n\nFor a meaningful revision, optionally compare drafts directly:\n\n```sh\nsst --format agent --compare BEFORE.txt AFTER.txt\n```\n\nRead the changed findings in context rather than the score delta. Stop when\nthe prose is better. Do not chase a target score, zero findings, or endless\npasses.\n\nReturn the improved text when revision was requested. Summarize only the\neditorial choices that help the user; mention important findings deliberately\nretained when that explains the result.\n",{"name":5,"description":7},"main","88cd6b591710e45b26c98f7af98381ac0220efc3","cmd\u002Fsst\u002Fskill",0,"https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool","https:\u002F\u002Fgithub.com\u002Flemon07r","2026-11-03 02:56:45",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add lemon07r\u002FSlopScoreTool --skill review-writing-slop",{"tool":37,"command":38},"bunx","bunx skills add lemon07r\u002FSlopScoreTool --skill review-writing-slop",{"tool":40,"command":41},"pnpm","pnpm dlx skills add lemon07r\u002FSlopScoreTool --skill review-writing-slop","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool\u002Fblob\u002Fmain\u002Fcmd\u002Fsst\u002Fskill\u002FSKILL.md",true,[45],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"agents\u002Fopenai.yaml","REFERENCE","参考文档",218,"https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool\u002Fblob\u002Fmain\u002Fcmd\u002Fsst\u002Fskill\u002Fagents\u002Fopenai.yaml",[52,57],{"skillNo":53,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M79S63HR92TMS9Z47BPECCS5","https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Freview-writing-slop\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M79S5ZAK9YC2XAW9Z3074F7F","develop-slop-score","Implement, review, debug, test, benchmark, or modernize this repository's Go writing analyzer and sst CLI. Use for scoring and diagnostic logic, Unicode tokenization and source spans, calibration data, CLI behavior, structured output, embedded assets, or analyzer evaluation.","https:\u002F\u002Fgithub.com\u002Flemon07r\u002FSlopScoreTool\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fdevelop-slop-score\u002FSKILL.md",[],[]]