idea-finder
Skill1 starsChanged 52 days agoMITView on GitHub
Generate genuinely novel technical/product/research ideas within a domain by finding unsolved constraint intersections — not generic brainstorming. Use whenever the user wants new invention/patent/paper/product ideas in a technical space, asks "what's a gap nobody's filled here," or wants direction before committing to a project. Every idea produced must survive a "why hasn't this been done" gut-check before being presented. Do not use for open-ended creative brainstorming unrelated to technical/patent/research output — this is for ideas that need to hold up, not for volume.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- zhenya-vlasov/idea-findertheir own version, first seen Jun 9, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- zhenya-vlasov/idea-finder★ 17same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository32 other files it gives its agents.
- agent-system-designSkill
- api-contract-designSkill
- architecture-designSkill
- architecture-optimizationSkill
- ci-speedupSkill
- code-handoffSkill
- cross-ai-handoffSkill
- crypto-storage-auditSkill
- database-designSkill
- dataset-sourcingSkill
- distributed-debuggerSkill
- embedded-constrained-codingSkill
- freedom-to-operate-analysisSkill
- frontend-architectureSkill
- github-masterSkill
- ieee-paper-drafterSkill
- memory-leak-hunterSkill
- ml-model-selectionSkill
- ml-security-auditSkill
- mobile-platform-architectureSkill
- observability-designSkill
- patent-drafterSkill
- patent-novelty-evaluatorSkill
- performance-optimizationSkill
- project-scaffoldingSkill
- researcher-evaluatorSkill
- researcherSkill
- safety-compliance-scopingSkill
- scale-refactorSkill
- secure-code-reviewSkill
- slow-query-optimizerSkill
- test-architectSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Before generating anything, pin down:
- Distinctive: line 28
Avoid: "apply [trendy technique] to [domain]" with no specific mechanism named.
- Distinctive: line 37
No real reason — this is a red flag, not a green light.
- Distinctive: line 53
Don't inflate the list to look productive.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Idea Finder
Most idea generation produces plausible-sounding combinations of buzzwords ("AI-powered X for Y"). That's not a gap, it's a mad-lib. This skill finds ideas at the intersection of constraints that are individually well-understood but jointly unsolved — the kind of gap that survives someone asking "wait, doesn't X already do this?"
Step 1 — Define the search space precisely
Before generating anything, pin down:
- The domain and its actual boundaries (not "AI" — "on-device inference for battery-constrained sensor nodes").
- The target artifact: is this idea meant to become a patent, a paper, a product feature, or a project? The bar and the shape of a good idea differ for each — a patentable idea needs a novel mechanism, a paper needs a testable question, a product feature needs a user who's actually blocked today.
- Any constraints already fixed by the user's situation (hardware, budget, deadline, existing project it has to plug into).
Step 2 — Map what's known, briefly
List the 3-5 dominant existing approaches in this space and, for each, the specific limitation it has — not "it's not good enough" but the actual mechanism of failure (latency bound, doesn't generalize past X, requires connectivity, breaks under adversarial input, etc.). This is a lightweight pass — hand off to the researcher skill before committing real time to any idea that survives Step 4.
Step 3 — Generate ideas at constraint intersections, not combinations of trends
The reliable pattern for a real gap: technique A solves problem P but fails under constraint C; technique B handles constraint C but doesn't solve P; nobody has combined them because the combination is non-obvious, not because it's easy and just unattempted. Look specifically for:
- A solved problem that breaks the moment a specific real-world constraint from Step 1 is added (offline, adversarial, low-power, small-data, real-time).
- A known technique whose limitation is usually "solved" by throwing out a constraint the user's domain can't actually drop.
- A place where two adjacent fields each have half the solution and nobody's connected them (this is the highest-yield pattern for defensible patents).
Avoid: "apply [trendy technique] to [domain]" with no specific mechanism named. If the idea can't be stated as "X handles constraint C by doing [specific mechanism], which Y currently can't because [specific reason]," it's not specific enough yet — push on it.
Step 4 — Force the "why hasn't this been done" test on every surviving idea
For every candidate, name the most likely reason it doesn't already exist:
- Genuinely hard (technical blocker) — good sign, worth pursuing.
- Nobody needed it until recently (a dependency/market condition just changed) — good sign, worth pursuing, and worth stating what changed.
- Economically not worth it for anyone else, but is worth it here — situational, state why.
- It's actually already been done and the idea generator just didn't know — kill it now, flag for a
researchercheck before the user gets attached to it. - No real reason — this is a red flag, not a green light. Ideas with no explanation for why they're unclaimed are usually not as novel as they look; say so.
Output structure
# Idea candidates: [domain]
## 1. [Idea name]
- Gap it fills: [specific mechanism vs. specific limitation]
- Why it likely hasn't been done: [reasoned answer from Step 4]
- Feasibility read: [rough — what would have to be true for this to work]
- Next step before committing: [researcher check / patent-novelty-evaluator check / prototype]
(repeat, ranked by conviction — 3 well-reasoned ideas beat 15 generic ones)Don't inflate the list to look productive. If only two ideas survive Step 4, present two — that's the point of the filter.
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.

