li-repurpose

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: none for the install itself: no account, no API key, no paid service, and nothing else to install first. For using the skill afterwards, read this first: it runs on one long asset I supply myself — a transcript, article, newsletter, script or call summary that I paste or give you as text. If I give a YouTube URL instead, SKILL.md says to use a transcript tool only if one is already available in the session (SKILL.md names no tool and no service); if none is available, I paste the text myself. One honest limit, disclosed here before anything is installed: the full workflow as written in SKILL.md is not self-contained in this folder — assigning a hook formula to each post uses hooks.json from the separate li-post skill, and drafting a post goes through the separate /li-post and /li-human skills (see point 2). Those skills are not part of this install and I will decide about them separately; the extraction and week-planning core of this skill works from this file alone, but hook assignment and drafting as written need them. This skill never logs into LinkedIn, never asks for my LinkedIn password, cookies or session, uses no browser automation and no LinkedIn API, and posts nothing: it writes a plan and, later, drafts — I post to LinkedIn myself. There is no script and no code in this skill at all: it is a single instruction file. Install "li-repurpose" for me. It turns one long asset — a YouTube video, podcast, newsletter, blog post, transcript or client call — into a week of LinkedIn posts. It reads the whole asset first, then extracts rather than summarises: it pulls out every claim (a sentence that would start an argument), every number (figure, cost, duration, percentage), every story (a person, a scene and a cost), every mechanism ("the way this works is..."), every admitted mistake, and every line already quotable as-is, and it lists what it found, with counts, before writing anything — if the asset yields fewer than four items, it says the asset is thin instead of squeezing four thin posts out of it. Then it builds the week: each extract becomes one post that must stand completely on its own (the reader has not seen the source and never will; never "as I mentioned in my latest video"), a varied hook formula is assigned to each, and they are ordered with the strongest claim first, the story midweek and the mechanism post last. The output is a SOURCE / FOUND / WEEK plan in the format shown in SKILL.md, and posts are drafted only on request, one at a time (for example "write Tuesday"), each through /li-post and /li-human. Repository: https://github.com/avi691/linkedin-agent-skill (the skill is the whole skills/li-repurpose/ folder at the pinned commit: SKILL.md only, 1 file) Pinned version: install the code at commit e4ec4fb6f1b0e227eb9b14b31b735072db5df9cb exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/avi691/linkedin-agent-skill into a temporary folder, check out commit e4ec4fb6f1b0e227eb9b14b31b735072db5df9cb, and copy the skills/li-repurpose/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-19: [FINGERPRINTS] - SKILL.md: SHA-256 9524d3958f153d8be3d679a372c8cb5065e30653ed2a7bf4596e921b2baa39df If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the file: - The skill folder holds 1 file only: SKILL.md (2,720 bytes). There are no scripts, no code, no JSON data files and nothing executable in the folder, and nothing to run at install time or at use time. - The references outside the folder, exactly as written in SKILL.md, and what they do and do not require: (a) "Assign a hook formula from li-post/hooks.json to each" post — hooks.json is not in this folder: in the repository it is skills/li-post/hooks.json (9,414 bytes), the 21 hook formulas of the separate li-post skill. Without li-post installed, you do not have those formulas (SKILL.md's example output names only a few: #1 Contrarian, #17 Time Anchor, #9 Cold Open, #21 Direct Value), so that assignment step is limited to what I supply or what a separately installed li-post provides. (b) Drafting, when I ask for it, goes "each through /li-post and /li-human" — two separate skills of the same pack, not files of this skill and not part of this install. What they bring with them, from their own SKILL.md files in the same repository: /li-post reads ~/.claude/linkedin/voice.md if it exists (my local voice profile; the repository's templates/voice.md is the blank template for it, 1,425 bytes, which SKILL.md of li-repurpose never mentions), writes that file from three of my past posts if it is missing, and on my "yes" appends a finished post to ~/.claude/linkedin/log.md; /li-human runs two local Python scripts from its own folder (humanize.py and detect.py, reading its slop.json) that clean and score a draft entirely on my machine. Do not install li-post, li-human or any other skill as part of this install, and do not copy hooks.json, templates/voice.md or any other repository file into this skill's folder. (c) The YouTube case: if I give a YouTube URL "and there is a transcript tool available in the session, use it; otherwise ask them to paste the text" — SKILL.md names no transcript tool, no API and no service; if no such tool is already in the session, the fallback is that I paste the transcript myself. Apart from these three points, SKILL.md needs no file from elsewhere in the repository — not the other 10 skills, not the README. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. No script is run during installation, no key, password, cookie or account is asked for, and there is no telemetry in the folder — it contains no URL at all. - What happens at use time, disclosed here before it ever happens — using the skill is a separate step that only starts when I supply a long asset and ask for a repurposing plan: the skill itself makes no network call, calls no API, opens no browser, does not log into LinkedIn, does not read LinkedIn cookies or any session store, and posts, schedules and sends nothing anywhere — the finished posts, if I later draft any, are text I paste into LinkedIn myself. The only material that enters our conversation is the asset I choose to give you (which can be sensitive: a client call transcript or summary, a newsletter, an article — I decide what to paste) and it is processed by you, my agent, under your own account and terms; the skill sends none of it to LinkedIn, YouTube or any third-party service, and no other recipient is named in SKILL.md. The single conditional exception: if I give a YouTube URL and a transcript tool is already available in the session, that URL goes to that tool so it can fetch the transcript — a tool SKILL.md does not name, under that tool's own terms; if I do not want that, I paste the text instead. Drafting through /li-post and /li-human, if those separate skills are installed and I ask for a draft, follows their own disclosed behaviour (local voice profile and log under ~/.claude/linkedin/, humanizer scripts run locally, nothing uploaded by them either, per their SKILL.md files). - Do not ask for my LinkedIn password, cookies or session, do not log into or scrape LinkedIn or YouTube, do not install li-post, li-human or any other skill, do not fetch any transcript, and do not repurpose anything during this install: repurposing only starts later, when I supply an asset and ask for it. 3. When I say "yes, install", install it as a skill: copy the whole li-repurpose folder into the agent's skills directory, in a folder named "li-repurpose", keeping its 1 file (SKILL.md at its root). A manual clone followed by copying the skills/li-repurpose folder into ~/.claude/skills/ is the route described here; any other route is fine only as long as every installed file matches its fingerprint. 4. Verify with no network calls: frontmatter valid (name: li-repurpose), the 1 file in place with matching fingerprint. There is no script to run to verify. 5. Report what was installed, where, and what I still need to do myself (for example: supply the long asset myself when I want a week of posts — paste a transcript, article, newsletter, script or call summary, or give a YouTube URL knowing a session transcript tool, if one is available, would receive that URL, otherwise paste the transcript; decide separately whether to install the sibling skills li-post (for its hooks.json formulas) and li-human (for drafting and humanizing as SKILL.md writes it), knowing they are not included here; if I use li-post, set up my voice profile ~/.claude/linkedin/voice.md from templates/voice.md or from three of my own posts; post the finished drafts to LinkedIn myself, since this skill never posts; using the skill is a separate step that only starts when I ask for it, and the use-time handling disclosed above happens only then). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.