Creator research skills for Claude Code and Codex: see what's working on YouTube, Instagram, and TikTok,…
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: Agent 技能. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository kabo-sh/kabo-plugins;first catalogued 2026-10-09;last seen 2026-10-16。There is no official certification or quality rating here — we record sources and dates only.
Kabo — creator research skills for Claude Code and Codex
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
Website ·
Install ·
Free creator tools ·
Report an issue
Kabo is an AI partner for creators. This repository installs the Kabo plugin into Claude Code or Codex: ask a question in plain language and get a report built from public channel, account, and video data, with sources, dates, and limits stated.
**Free to use.** Every Kabo account gets a free daily allowance.
See it in action
The daily TikTok sound board — the board Kabo's TikTok trend research reads: how many videos use each sound and how many were added today. Captured from kabo.sh on 2026-10-09.
Breakouts against the channel's own baseline — a recorded views trend, and videos ranked by how far they beat the channel's typical video of the same format. Captured from kabo.sh/creators on 2026-10-09.
Outliers in the Kabo workspace — each video's views as a multiple of its creator's recent median, filterable by platform, format, and time window. Kabo runs in the browser too; this plugin brings its research into Claude Code and Codex. Captured from the Kabo web workspace on 2026-10-09.
What you can do
| Job | What Kabo does | Platforms |
|---|---|---|
| **Spot what's working now** | Scans current trends and the patterns behind high performers; reads the daily Instagram and TikTok trending-audio boards (usage counts, daily growth, fastest risers) and TikTok's rising search terms; surfaces emerging creators | YouTube · Instagram · TikTok |
| **Study competitors** | Finds comparable accounts, compares their recent content against each account's own baseline, and names positioning gaps | YouTube · Instagram · TikTok |
| **Review an account** | What works, what hurts, and what to continue, stop, or test; engagement rate; follower loss. Ask for a **deep review** and Kabo also watches the account's videos — frames and transcript — at every step | YouTube · Instagram · TikTok |
| **Diagnose a reach drop** | Works out whether a drop, stall, or audience mismatch is real and where it comes from | YouTube · Instagram · TikTok |
| **Break down a video** | Timestamped diagnosis of the hook, pacing, and retention risk — for your video, someone else's, or an unpublished cut on your machine — plus how it was edited | Any public video link or local file |
| **Decide what to make next** | Content ideas from your account, trends, competitors, and comments, then scripts, hooks, and titles for the one you pick | YouTube · Instagram · TikTok |
| **Plan when to post** | A four-week content calendar (CSV/ICS export) and posting times drawn from your own history — never a generic "best time" chart | YouTube · Instagram · TikTok |
| **Build your brand** | Channel names, bios, and usernames within each platform's limits, a profile visual spec, and a media kit for brand outreach | YouTube · Instagram · TikTok |
| **Make money** | Eligibility and threshold gaps for each platform's monetization programs; line-item pricing and a negotiation reply for a brand-deal brief | YouTube · Instagram · TikTok |
| **Check a platform rule** | Verifies rules, thresholds, features, and rumours against official help pages, with quotes, links, and retrieval dates | YouTube · Instagram · TikTok |
New here? Run /kabo-start: a short questionnaire, one real analysis of your own account, a 90-day plan, and your first piece of content.
Example requests
How it works
1. **You ask in plain language.** Kabo picks the matching research skill and shows it to you before anything runs.
2. **Kabo fetches the public data.** Data is collected on Kabo's servers, so there are no API keys to set up. Every skill is signed by the platform and verified on your machine before it executes.
3. **You get a report with its evidence.** Findings come with source links, the time window, the sample size, and what could not be measured.
What Kabo does not do:
- **It does not publish for you.** It analyzes, plans, and drafts; posting stays with you.
- **It does not guess private metrics.** Click-through rate, retention, revenue, and Insights are never inferred from public views or likes.
- **It does not promise virality.** When the evidence is thin, the report says so.
How Kabo compares
We make Kabo, so read this as our view. Every cell comes from each product's own public pages, checked on 2026-10-09. If something has changed, please open an issue.