Menu
Develops ONE chosen direction into a shaped concept — the core idea, how it works, what it needs, the open questions. Use to deepen a single idea, or /ideate. Not for generating many (brainstorm) or deciding among them (propose).
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | ideate |
| description | Develops ONE chosen direction into a shaped concept — the core idea, how it works, what it needs, the open questions. Use to deepen a single idea, or /ideate. Not for generating many (brainstorm) or deciding among them (propose). |
The depth skill. Given one chosen direction (often a pick from a brainstorm shortlist), it develops it
into a shaped concept: what the idea actually is, how it would work, what it depends on, what is still
unknown. It goes deep on a single idea; it does not generate many (that is brainstorm) or choose among
options (that is propose).
It is read-only and advisory: it writes a CONCEPT.md for the one idea, never product code. The output is
judged by Review — is the concept concrete enough to decide on or prototype, and are the open questions
named honestly rather than hidden?
Does: take one direction; articulate the core idea in a sentence; work out how it functions (the mechanism,
the shape, the primary flow); name what it depends on and assumes; surface the open questions + risks; sketch
a first-cut of what "good" looks like; write CONCEPT.md + a receipt. Does not: generate alternatives (that
is brainstorm), pick between concepts (that is propose), build a proof-of-concept (that is prototype),
or write product code.
brainstorm first.propose.prototype.specify.brainstorm shortlist item, or a stated idea) + its intent.state/project-map.md, constraints, the intake/spec if one exists.specify).CONCEPT.md — core + mechanism + dependencies/assumptions + open questions + a sketch of good.concepts/CONCEPT.md if that dir exists) — persistence:
Ephemeral. Core (one sentence) · mechanism · dependencies + assumptions · open questions + risks ·
sketch of "good".<workspace>/receipts/ideate-receipt-<ts>.json): base schema + concept, core_statement,
dependencies, open_questions, concept_path. persistence: Ephemeral.Done when: CONCEPT.md develops one direction into a concrete concept — a sharp core sentence, a
mechanism, its dependencies/assumptions, honestly-named open questions, and a sketch of "good"; a receipt
records the concept + open questions.
Not done if: it generated alternatives instead of developing the one (that is brainstorm); the core
sentence is still fuzzy; or the open questions/risks were hidden to make the concept look more settled than
it is.
Next step: propose decides between this and other concepts; prototype de-risks an open feasibility
question; specify turns a settled concept into a task card.
ideate develops ONE of them in depth.ideate produces the concept it weighs.ideate names the question.Divergent idea generation — many candidate ideas across several frames with judgment deferred, then a clustered shortlist. Use to open a solution space wide, or /brainstorm. Not for developing one idea (ideate) or picking one (propose).
Structures a presentation narrative — the story arc, a slide-by-slide outline, and speaker notes — from results or receipts. Use to shape a talk or deck, or /present. Not for prose reports (report) or rendering visuals (asset).
Converges to a recommended approach — the options, their tradeoffs against stated criteria, a pick, and the rationale (plus what would change the call). Use to decide among approaches, or /propose. Not for generating options (brainstorm) or deep-diving one (ideate).
Plans a throwaway proof-of-concept to answer ONE feasibility unknown — the question, the smallest build that answers it, the success signal, and the discard plan. Use to de-risk before committing, or /prototype. Not for the real build (executor) or deciding (propose).
Produces a REDTEAM-FINDINGS.md probing your OWN AI app for jailbreak/injection/exfil/tool-abuse susceptibility, each reproduced and paired with a hardening, after a blocking authorization+ownership check. Use to red-team an owned AI system defensively.
Produces a PENTEST-PLAN.md then PENTEST-FINDINGS.md for your OWN app's attack surface (each finding reproduced, non-destructive, with remediation) after a blocking authorization+ownership check. Use to test an owned app's attack surface defensively.