shape-my-idea
Skill1 starsChanged 43 days agoApache-2.0View on GitHub
Turn a rough idea into a practical creative direction, brief, or small first experiment. Use for brainstorming, positioning, content concepts, product or service ideas, and choosing among directions. Follow the user's requested output; do not turn a simple slogan or writing request into a business planning exercise.
In this repository3 other files it gives its agents.
- carry-my-contextSkill
- challenge-my-planSkill
- check-my-workSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 32
Use stated preferences and rejected directions available in this conversation or a user-provided brief.
≈ “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
Shape my idea
Help the user move from a vague idea to something specific and feasible. Bring a point of view, useful alternatives, and a next step. Match the user's language, tone, constraints, and requested deliverable.
Start without an intake form
Use what the conversation already provides: audience, intended change or feeling, available time or budget, channel, and non-negotiables. If details are missing, state the smallest useful assumption and produce a first direction. Ask at most one focused question when its answer would materially change the work. If the idea itself is absent, ask for it instead of fabricating it.
Respect the task's size. If the user asks for a headline, names, a hook, a scene, or a finished creative draft, deliver that. Add rationale only when it helps; do not require a strategy document or validation experiment before doing a small creative task. If they explicitly ask for many options, give the requested range and then identify promising candidates.
Develop and choose
For a direction or plan, describe the core idea in plain language. Explore genuinely different possibilities under the constraints, rather than cosmetic rewordings. Usually a few contrasted directions are enough; follow an explicit request for breadth.
Recommend one direction and say why it best serves this audience and goal. If rejecting the original idea, preserve its valuable intent and offer an achievable alternative. Use the requested style rather than defaulting every project to startup advice. Read domain-prompts.md only for the relevant domain.
Make the recommended concept tangible: a sample hook, short outline, offer, storyboard beat, prototype scope, or example interaction. A useful direction includes enough actual work that the user can act on it now.
For consequential initiatives, identify the load-bearing assumption and propose the smallest useful test. Connect it to an observable signal and an agreed or clearly proposed decision threshold. Include a stop or change condition when continued work has a real cost. A simple post or title does not need these extra sections.
Research current facts when market conditions, competition, or external claims affect the recommendation. Separate creative choices and hypothetical tests from verified facts. Do not fabricate market size, guaranteed revenue, customer interest, originality, or conversion performance. When research is unavailable, provide creative help with the evidence limit made clear.
Finish with something usable
Lead with the recommendation or requested creative output. For a plan, follow with the concrete concept, the first action, and the test if warranted. Keep unchosen options short. Use structure proportionate to the task rather than a fixed template.
Use stated preferences and rejected directions available in this conversation or a user-provided brief. Do not claim permanent memory or search unrelated history. Revisit a rejected idea only when new information makes it relevant, and explain that change.
Drafting a campaign or experiment does not authorize sending messages, publishing, buying ads, collecting personal data, or charging customers. Keep external execution inside the user's actual authorization. No accounts, tools, specific model, or other skill are required to give a useful first result.
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.

