A personal LinkedIn scheduler is small: a draft editor, a queue, LinkedIn's Share API to publish to your own profile, and a language model to rewrite in your voice. That is a weekend, most of it spent on the developer app and OAuth. What Taplio sells beyond that sits outside any official API: a library of millions of viral posts, a lead database, reply and DM automation that acts as you, and stats collected through an extension. LinkedIn's rules ban third-party automation and scraping, so a personal build should leave those parts out.
Before you buildBuffer's free plan schedules 10 posts per channel to LinkedIn at no cost. If you just want a queue, try that first.
Build me a LinkedIn post studio to replace Taplio for my own profile. Requirements: - Node.js 22 + TypeScript: a Fastify server and a React editor built with Vite. `npm start` serves http://localhost:4730 behind one password from .env. - Storage: better-sqlite3 in ./data/posts.db: drafts (text, status, scheduled_at, published_at, LinkedIn post URN), an ideas inbox, voice samples (my past posts), and analytics rows (post, date, impressions, reactions, comments, shares). - LinkedIn: OAuth 2.0 with the Share on LinkedIn product (scope w_member_social, plus openid and profile to get my person URN). CLIENT_ID, CLIENT_SECRET and REDIRECT_URI live in .env; tokens are stored encrypted with APP_SECRET. Publish text, link and single-image posts through the UGC Posts API (POST /v2/ugcPosts) and stay under the documented 150 member requests a day. - Scheduler: `croner` checks every minute and publishes drafts that are due. A failed publish retries 3 times with backoff, then marks the draft failed with the error text. Nothing publishes unless I pressed Schedule on that draft. - Editor: live count against the 3,000-character limit, a feed-style preview, and a button that turns selected text into Unicode bold or italic. - AI assist: Anthropic (ANTHROPIC_API_KEY) or any OpenAI-compatible endpoint from .env. Actions: ten ideas from a topic, rewrite in my voice using five voice samples as examples, shorten, and three stronger first lines. Results appear as suggestions and never replace my text on their own. - Analytics: import the XLSX that LinkedIn's post analytics page exports (`xlsx` library), map columns by header name, and chart impressions per post, the weekly trend and the best posting hour. - Single user. No accounts, no telemetry, local-first. Outbound calls: LinkedIn's API and the LLM endpoint. - Out of scope: scraping LinkedIn, a viral post library, lead lists, automatic comments or DMs, connection requests (LinkedIn's terms ban automation and scraping). - README: creating the LinkedIn developer app and adding the Share product (budget an hour), how to re-authorise when the token expires, and the steps to export analytics from LinkedIn.
Opens your agent with the prompt filled in, you press enter. Or copy it and paste it into an empty folder.
The first prompt builds the core. Run these one at a time in the same folder to grow it into your own Taplio.
Add a carousel maker to the LinkedIn post studio in this folder. Requirements: - From a draft, ask the LLM endpoint for 6 to 10 slides of at most 25 words each (JSON in, JSON out) and show them as editable cards. - Render the slides to a 1080x1080 PDF with Playwright from an HTML template; offer two themes and a logo upload, and keep the PDF in ./data/carousels/<draft id>.pdf. - Add a download button. Do not upload to LinkedIn automatically: I attach the PDF myself. - Escape every slide text in the template. - Update the README with the Playwright install step.
Add a weekly review to the LinkedIn post studio in this folder. Requirements: - Every Monday at a configured hour, build a digest from the imported analytics: top 3 posts by impressions, the post with the best comment-to-impression ratio, and posting days I skipped. - Add a one-paragraph note from the LLM endpoint that compares the week with the previous four weeks, and say clearly when there is too little data to conclude anything. - Send the digest through SMTP from .env and keep a copy on a Reviews page. - Update the README with how to schedule it.
Add an idea importer to the LinkedIn post studio in this folder. Requirements: - Watch a folder (NOTES_DIR in .env) for .md and .txt files and RSS feed URLs listed in a feeds.txt file; add new items to the ideas inbox with their source. - On a button, send one idea to the LLM endpoint and create a draft that uses my voice samples; mark the draft as AI-assisted in the database. - Deduplicate by content hash and let me archive or pin ideas. - Update the README with the folder layout.
Solo creators and founders want one place to find ideas, write, schedule and engage, with an inspiration library and analytics LinkedIn does not give them. Reviewers on Reddit are mixed: one says it is good for finding viral posts but weak at writing, and names AuthoredUp as better for drafting.
Estimate · LinkedIn's Share API is free. Running costs are a few dollars of LLM tokens a month and, if the laptop is not always on, a $5 VPS. The build is token usage on your agent plan.
Checked 2026-10-19 against Taplio's docs: 1, 2, 3, 4, 5, 6, 7
| Plan | Monthly | Annual, per month | What you get |
|---|---|---|---|
| Starter | $39 | $29.25 | Post scheduling, drafts and core analytics. The compare table caps AI credits at 250 and comment credits at 500 a month on the lower plans. |
| Growth | $69 | $51.75 | Everything in Starter plus AI viral post writing, faster replies and comments, and team collaboration. |
| Pro | $199 | $149.25 | Everything in Growth plus unlimited AI posts and comments, lead outreach automation, a 90-day auto-reply monitoring window and priority support. |
No free plan: every plan starts with a 7-day trial that gives full Pro access, and the Chrome extension works without a plan.
Kinda. The core is a weekend build with the prompt on this page, but real gaps remain: The library of 5M+ viral posts and the monthly LinkedIn Benchmark, Lead lists, automatic DMs and connection requests.
About $39 a month on the Starter plan per seat, $468 a year (checked 2026-10-19).
The library of 5M+ viral posts and the monthly LinkedIn Benchmark; Lead lists, automatic DMs and connection requests; Suggested replies and comments on other people's posts, plus the Chrome extension; Stats pulled straight from your profile instead of an exported file; The AI carousel and image generators. If any of those matter to you, keep paying.
Yes: Postiz (open source), BrightBean Studio (open source), Buffer Free (free).