YominPost

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. The installed skill is a single instruction file. Using it by hand needs nothing else; the optional YominPost tool (MCP or CLI), and uv for its zero-install route, only come in at use time, if I choose them then — do not install or run them during this install. Install "yominpost" for me. It is a social-media writing playbook: a single SKILL.md that turns one piece of material into one native post per platform, runs a pre-publish platform-risk check on each draft, and hands the drafts to me for approval — it never publishes without the user. Repository: https://github.com/mrlong0129/yominpost (the skill is the whole skill/ folder at the pinned commit: SKILL.md only) Pinned version: install the code at commit 12d7a35ea5ed3b10ba132df76f1029fa846c6ed0 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/mrlong0129/yominpost into a temporary folder, check out commit 12d7a35ea5ed3b10ba132df76f1029fa846c6ed0, and copy the skill/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-19 and reverified on 2026-10-23: [FINGERPRINTS] - SKILL.md: SHA-256 9b441bd23048507cb978811ed3ab00b86ebef9f0add24abda412da40d5f0d1da 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 files: - The skill folder holds one file only: SKILL.md (3,652 bytes). There is no code, no script and no dependency list in the folder: the skill is a writing and safety-check playbook (per-platform formats, native rewriting rules, one self-critique pass, a pre-publish risk check, and a hand-off step). - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. SKILL.md itself names no account, no API key, no telemetry and no remote service: the only hosts it names are github.com (the use-time uvx route below) and 127.0.0.1:8300 (the local YominPost studio). - At use time, the playbook itself works on material I provide: pick the format per platform first (X text ≤ 280 characters or a thread, LinkedIn 600–1,300 characters, Instagram/Xiaohongshu image post or carousel, TikTok/YouTube Shorts short-video storyboard, Telegram/Discord/Mastodon/Bluesky text), rewrite natively for each platform without inventing facts, numbers, customers or quotes, self-critique once, then run the pre-publish risk check on every post: block and fix prompt leaks, engagement bait and over-length posts, and warn on more than about 5 hashtags, more than 1 link, repeated posting of the same link or near-identical text, or a brand-new account posting several times a day. - The hand-off is the last step, always: show the drafts with the format reason and the risk-check result, save them as drafts or give them to me to post. The skill says never to publish without the user, and never to claim something was published unless I did it, or explicitly asked and a real publishing tool did it. - If the YominPost tool is available, the skill uses it at use time: MCP generate_posts(brand, material, platforms, num_topics), then check_post(text, platform) on each candidate, then save_draft(...) for the ones I like, with review at http://127.0.0.1:8300 (local); or the CLI: yominpost run --brand "<brand>" --source "<material>" --platform x --platform linkedin --json. Its default template driver is an offline scaffold; the skill suggests YOMINPOST_PROVIDER=claude_code for model-written copy. - If YominPost is not installed but uv is available, SKILL.md proposes a zero-install route at use time: uvx --from git+https://github.com/mrlong0129/yominpost yominpost run ... --json, which downloads the full YominPost application from GitHub at that moment (that route is not pinned to the commit above in SKILL.md, and the full application is outside the skill/ folder). Do not run it during this install: it is my decision, later, if I ever ask. Credentials for optional providers or platform accounts exist only in the full application's own documentation, outside the skill folder; none is part of this install, and none should be requested during it. 3. When I say "yes, install", install it as a skill: copy the whole skill folder into the agent's skills directory, in a folder named "yominpost", keeping its file (SKILL.md). The upstream route, downloading skill/SKILL.md from the repository into ~/.claude/skills/yominpost/ (or ~/.cursor/skills/yominpost/), installs the same single file; either route is fine as long as the file matches the fingerprint. 4. Verify with no network calls: frontmatter valid, file in place. 5. Report what was installed, where, and what I still need to do myself (for example: provide the material for any post myself; using the skill is a separate step that only starts when I ask for it, and the use-time routes disclosed above — the local YominPost tool, or the uvx download of the full application — happen only then, if I choose them). 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.