video-ideation
Skill91 starsChanged 11 days agoMITView on GitHub
Use when the user wants video ideas for a YouTube channel, what to make next, or topics for the coming month. Requires the TubeLab MCP.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audit-channel-riskSkill
- channel-analysisSkill
- comment-miningSkill
- competitor-researchSkill
- monetization-researchSkill
- niche-bendingSkill
- niche-finderSkill
- niche-validatorSkill
- niche-watchlistSkill
- reverse-engineer-channelSkill
- script-breakdownSkill
- sponsorship-prospectingSkill
- thumbnail-conceptsSkill
- title-formulasSkill
- trend-reportSkill
- tubelab-mcpSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
Keep at most 2 winners per other channel.
- Distinctive: line 56
If it runs long, cut whole items.
- Very specific: line 63
Do not search without publishedAtFrom and classificationQuality at "neutral" or above, or the results fill with low-effort videos.
- Distinctive: line 65
Do not trust ratios from channels with a few hundred subscribers.
- Distinctive: line 66
Do not mix in Shorts, and do not pitch a video the channel already made.
- Distinctive: line 67
Do not write titles over 60 characters.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Video Ideation
Overview
About ten ideas for one channel, drawn from three sources and ranked: the channel's own winning formats, outliers in its niche, and related niches. A fourth source, channels with the same audience, runs only on request. Every idea needs proof that the audience wants it (potential) and proof that this channel can deliver it (fit). An idea with demand but no fit belongs on someone else's channel. Adapt each winner with a twist; never copy it.
Workflow
- Clarify scope only if needed. Unless told otherwise, the goal is reach, or conversion if the channel sells something.
- Source one, the channel's own formats. Read the channel and find the formats that win for it by ratio. Double down on each with a new twist. A new channel has no formats of its own: say so, and lean on the other sources.
- Source two, outliers in its niche. Search them and twist the best for this channel.
- Source three, related niches. Pick one or two neighbouring niches, find the formats winning there, and twist them.
- Only if the user asks for ideas from channels with the same audience: run one channel search with gender, age band and region plus a subscriber floor, with no retry. Assume the profile from who the channel addresses, unless a channel-search hit gives it, and say so in the Source bullet. If fewer than two usable channels come back, say so.
- Keep at most 2 winners per other channel; the channel's own formats have no limit. Fetch and view the thumbnail of every source video, and say so if one fails to load. Expect about three to four searches by default.
- For the twists you may use SCAMPER: substitute, combine, adapt, modify, put to another use, eliminate, or reverse. It is a tool, not a form to fill in.
- Judge each idea on potential (how strong the demand evidence is) and fit (how well it suits what this channel already wins with). Rank on both.
Output Format
# 💡 Video Ideas: [{channel}]({url})
- **Best idea:** {title, and why in one sentence}
## 🏆 Ranking
Potential and fit are judgment calls: High, Medium or Low.
| # | Idea | Source | Potential | Fit |
|---|---|---|---|---|
## 💡 Ideas
### 1. {Title}
[]({video url})
- **Source:** {own format, niche outlier, related niche, or same audience (on request, noting an assumed profile)}
- **Proof:** [{channel}]({url}) · [{title}]({video url}) · {views, ratio, date}
- **Twist:** {the SCAMPER move and what it changes, in one sentence}
- **Why it fits:** {one sentence}- Length: about 1,100 words. About ten ideas, with every source used represented, best first.
- Language: plain words and short, complete sentences. Nothing compressed.
- Bullets: one fact each, under 25 words, except quotes and verbatim titles. No extra sections or bullets.
- If it runs long, cut whole items. Never cut proof, and never squeeze sentences.
- Links: link every video, channel, moment, brand and offer you cite, built from IDs, handles and links in the data. Never guess one.
- Own formats: for an idea from the channel's own format, the proof is the channel's own video, and the linked title is enough.
- Thumbnails: use the 320px
mqdefault.jpgimage, on its own line under the heading, linked to the video.
Common Pitfalls
- Do not search without
publishedAtFromandclassificationQualityat"neutral"or above, or the results fill with low-effort videos. - Do not take proof from another audience or language, from sister channels, or from paid placements (huge views, almost no likes). Outlier hits carry no audience data, so judge it from the title and thumbnail; the language tag can call a Hindi or Urdu video English.
- Do not trust ratios from channels with a few hundred subscribers.
- Do not mix in Shorts, and do not pitch a video the channel already made.
- Do not write titles over 60 characters.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

