[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-96174":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":17,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":16,"starSnapshotCount":16,"syncStatus":34,"lastSyncTime":35,"discoverSource":36},96174,"linkedin-agent-skill","Jakeschincariol\u002Flinkedin-agent-skill","Jakeschincariol","Eleven free Claude skills that run a LinkedIn account: posts off 21 hook formulas, comments, replies, profile score, weekly plan, and a humanizer that strips the AI fingerprint and scores the draft before it goes out.","https:\u002F\u002Fopusjake.ai\u002Fr\u002Flinkedin-agent",null,"Python",135,21,102,1,0,29,46.93,"MIT License",false,"main",true,[24,25,26,27,28,29,30],"agent-skills","ai-humanizer","claude","claude-code","claude-skills","content-marketing","linkedin","2026-09-20 04:01:32","# The LinkedIn agent skill\n\nEleven Claude skills that run a LinkedIn account. Free, MIT, no signup, no API\nkey, nothing to connect.\n\nOne of them writes your posts off 21 hook formulas. One comments on other\npeople's posts. One handles the replies under yours. One scores your profile\nout of 100 and rewrites what lost points. One plans the week: what to post,\nwhen, and who to engage with.\n\nAnd one is the humanizer, which is the reason the rest are usable. It strips\nthe em dashes, the slop vocabulary and the invisible watermark characters out\nof a draft, then scores what is left against a five-check detection panel\nbefore you ever see it.\n\n**Nothing gets posted until you say yes.** These skills write. You post.\n\n## Install\n\nPaste this into Claude:\n\n```\nhttps:\u002F\u002Fgithub.com\u002FJakeschincariol\u002Flinkedin-agent-skill\n\nInstall this skill, then confirm \u002Fli-post works.\n```\n\nOr do it yourself, in Claude Code:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FJakeschincariol\u002Flinkedin-agent-skill.git\ncp -r linkedin-agent-skill\u002Fskills\u002Fli-* ~\u002F.claude\u002Fskills\u002F\n```\n\nOr as a plugin:\n\n```\n\u002Fplugin marketplace add Jakeschincariol\u002Flinkedin-agent-skill\n\u002Fplugin install linkedin-agent\n```\n\nProject-local instead of global: copy the same folders into your repo's\n`.claude\u002Fskills\u002F`. No Claude Code at all? Paste any single `SKILL.md` at the\ntop of a chat and it runs as a mode - you lose the two Python tools, which is\nmost of the point of `\u002Fli-human`, but the rest works.\n\nThen spend ten minutes on `templates\u002Fvoice.md`. Copy it to\n`~\u002F.claude\u002Flinkedin\u002Fvoice.md` and fill it in, or paste three of your own posts\ninto Claude and say \"write my voice.md from these\". Every skill reads that\nfile. Skip it and everything comes out sounding like everyone else.\n\n## The eleven\n\n| command | what it does |\n| --- | --- |\n| `\u002Fli-post` | One idea into a post. Three hook options from [21 formulas](skills\u002Fli-post\u002Fhooks.json), one full draft, humanized before you see it. |\n| `\u002Fli-comment` | Comments on other people's posts. Nine types, picked by what the post actually is. Never \"Great post!\". |\n| `\u002Fli-reply` | The thread under your own post. Sorts every comment into lead \u002F substance \u002F peer \u002F support \u002F noise, then writes in that order. |\n| `\u002Fli-profile` | Scores your profile against a [12-part rubric](skills\u002Fli-profile\u002Frubric.json) out of 100, then rewrites in fix-first order. |\n| `\u002Fli-plan` | The week. What to post, when to post it, and the 10 people to engage with. Writes `~\u002F.claude\u002Flinkedin\u002Fplan.md`. |\n| `\u002Fli-human` | The humanizer. Two scripts that actually run. See below. |\n| `\u002Fli-carousel` | Document posts. Slide-by-slide copy, the cover that earns the swipe, and the PDF to upload. |\n| `\u002Fli-repurpose` | One video, newsletter or transcript into a week of posts that each stand alone. |\n| `\u002Fli-dm` | The 200-character invite note, the first message, and the two follow-ups. Two. |\n| `\u002Fli-inbox` | Triages the inbox into lead \u002F recruiter \u002F peer \u002F ask \u002F spam, and tells you which tell gave the sequence away. |\n| `\u002Fli-audit` | Post-mortem on what you have already published. Ranks by engagement rate and reach multiple, not impressions. |\n\n## The humanizer\n\n`\u002Fli-human` ships two Python scripts with no dependencies. They run on your\nmachine, on your text, and nothing is uploaded.\n\n```bash\npython3 humanize.py draft.txt --report      # clean it, show every change\npython3 detect.py draft.txt                  # score it, five checks\npython3 detect.py before.txt after.txt       # prove the delta\n```\n\n**What comes out automatically:**\n\n- **Invisible characters.** Zero-width spaces and joiners, word joiners, soft\n  hyphens, byte-order marks, Unicode tag characters, non-breaking and narrow\n  spaces. Your keyboard does not make these. They survive copy-paste and they\n  are invisible in every editor you own.\n- **Typography.** Em dash to comma, en dash to hyphen, curly quotes to\n  straight, ellipsis to three dots.\n- **The lexicon.** 113 stock words and phrases with plain-English\n  replacements - delve, leverage, robust, seamless, crucial, testament to, \"in\n  today's fast-paced world\", \"let that sink in\" - with capitalisation preserved\n  and URLs untouched. It lives in\n  [`slop.json`](skills\u002Fli-human\u002Fslop.json) and it is meant to be edited.\n\n**What gets flagged instead of fixed:** \"It's not just X, it's Y\", rule-of-three\ntriads, one-word rhetorical questions, hashtag walls, reflex engagement bait,\nuniform sentence length. Changing the shape of a sentence needs judgement, so\nthose are handed back for a rewrite rather than mangled by a regex.\n\n**The five checks**, scored 0-100, higher is more human:\n\n| check | what it measures |\n| --- | --- |\n| BURSTINESS | sentence-length variation. Models write even. |\n| SPECIFICITY | numbers, names and concrete markers per 100 words |\n| SLOP DENSITY | lexicon hits per 100 words |\n| FINGERPRINT | invisible characters, em dashes, curly quotes per 1,000 |\n| VOICE | contractions, person, structural tells |\n\nThe verdict weights the mean at 60% and the **weakest single check** at 40%,\nbecause a detector only needs one signal to fire.\n\nRun against a deliberately terrible draft:\n\n```\n  BURSTINESS    ##################......  73.0\n  SPECIFICITY   ######################## 100.0\n  SLOP DENSITY  ........................   0.0    19 stock terms, 24.1 per 100 words\n  FINGERPRINT   ........................   0.0    1 invisible, 1 em dash, 3 curly quote\n  VOICE         ########................  33.3    3 structural tells\n  ------------------------------------------------------------\n  HUMAN SCORE   ######..................  24.8   FLAGGED\n```\n\nAfter `humanize.py`, with the flagged structures still unrewritten:\n\n```\n  HUMAN SCORE   #################.......  69.7   REVIEW    (+44.9)\n```\n\nThe last stretch to PASS is the part the script deliberately leaves to you.\n\n## The fine print, which is the honest part\n\n**These skills do not post to LinkedIn, and they should not.** There is no\nofficial API for posting to a personal profile without an approved partner\napp, and automating the site with a browser or a third-party tool violates\n[LinkedIn's User Agreement](https:\u002F\u002Fwww.linkedin.com\u002Flegal\u002Fuser-agreement) and\ngets accounts restricted. So every skill here ends the same way: a copy-ready\nblock, and you paste it. That is not a limitation bolted on afterwards, it is\nthe design. It is also why the approval gate is real rather than a setting.\n\n**The five checks are local heuristics, not detector APIs.** They are modelled\non the signals public detectors key on, and they run entirely on your machine.\nThey are not GPTZero, Originality, Copyleaks, Winston or Turnitin, they do not\ncall those services, and they cannot promise those verdicts. Fixing what they\nmeasure tends to move those numbers, because they are measuring the same\nunderlying things. That is the whole claim. Nobody can honestly sell you\n\"undetectable\", and anybody who does is selling you something.\n\n**The invisible-character pass is real and it is narrow.** It removes the\nzero-width and format characters that end up in generated text and survive a\ncopy-paste. That is a genuine, checkable fingerprint. It is not a claim about\ndefeating a cryptographic watermarking scheme, and this repo does not make\none.\n\n**Nothing here fabricates.** No invented metrics, clients or outcomes go under\nyour name. If a draft needs a number you have not given, it comes back with\n`{{your number}}` in it and a flag, every time.\n\n## Files\n\n```\nskills\u002Fli-post\u002Fhooks.json        21 hook formulas: template, example, what it is for, how it gets ruined\nskills\u002Fli-human\u002Fslop.json        the lexicon: 113 terms, 17 invisible classes, 11 structural tells\nskills\u002Fli-human\u002Fhumanize.py      the three cleaning passes\nskills\u002Fli-human\u002Fdetect.py        the five-check panel\nskills\u002Fli-profile\u002Frubric.json    the 100-point profile score\ntemplates\u002Fvoice.md               your voice profile. Fill this in first.\n```\n\n## Credit\n\nMade by Jake Schincariol, [opusjake.ai](https:\u002F\u002Fopusjake.ai).\nThe full write-up is at [opusjake.ai\u002Fr\u002Flinkedin-agent](https:\u002F\u002Fopusjake.ai\u002Fr\u002Flinkedin-agent).\n\n## License\n\nMIT. Take it, change it, ship it.\n","这是一个为 Claude AI 设计的开源 LinkedIn 运营辅助工具集，提供 11 个免配置、零依赖的技能（skills），覆盖内容创作、互动管理、个人资料优化与工作流规划等环节。核心功能包括基于 21 种钩子公式的帖文生成、智能评论与回复、Profile 评分与重写、周度发布与互动计划，以及关键的「AI 人类化」功能——自动清除 AI 文本特征（如冗余标点、模板化措辞、隐式水印），并进行五维检测评分。所有输出均需人工确认后才可发布，强调人机协同与合规可控。适用于 LinkedIn 个人品牌建设、B2B 营销、职场影响力提升等轻量级、高自主性运营场景。",2,"2026-09-11 02:30:11","CREATED_QUERY"]