Files
magnus919_agent-skills/genius-life/references/development-mode.md
Magnus HedemarkGitHubfactory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
9a42be585e feat(skill): add genius-life creativity practice skill (#299)
* feat(skill): add genius-life core skill files

Add genius-life/SKILL.md and the six references/ files (creative-process,
practice-mode, development-mode, practices-catalog, evidence-basis,
scope-and-safety) as original synthesis from the mission research library,
with copyright-compliant paraphrase, named-fellow attribution, and honest
framing. Templates, README, evals, and catalog integration ship in later
features.

Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>

* feat(skill): add genius-life templates, README, and evals

Add six fillable worksheets (talent audit, session plan, project
worksheet, conditions audit, incubation log, risk and failure review),
a human-facing README with the repository's required sections, and a
12-case eval manifest covering both modes and the required boundary
behaviors.

Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>

* feat(skill): add genius-life to catalog and regenerate artifacts

Add the genius-life README catalog entry at its sorted position and
regenerate llms.txt, .claude-plugin/marketplace.json, .codex-plugin/plugin.json,
and .agents/plugins/marketplace.json from their generators.

Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>

---------

Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
2026-08-08 01:31:20 -04:00

6.8 KiB

Development Mode — Structured Idea Development

Development mode is for a concrete project, idea, or problem the person wants to move forward. It runs the shared five-phase spine explicitly — Prepare → Immerse → Incubate → Capture → Verify — with worksheet support and user-owned decisions throughout. The agent structures the process; it does not decide for the person. The shape of the process comes from the unified picture Denise Shekerjian assembled in Uncommon Genius (1991) and from the modern research summarized in Evidence basis.

Mode selection

  • Infer the mode from the request. A concrete project, idea, or problem to move forward means Development mode. An open, aspirational request ("help me be more creative," a block, habits) means Practice mode.
  • Ask one compact clarifying question only when the request is genuinely ambiguous between the two. Do not default to asking, and do not run a long intake interview. Example of a legitimate question: "Do you want to develop this specific idea, or work on creativity more broadly?"
  • If the person shifts modes mid-session, re-agree on the work in one or two sentences and re-read the relevant reference before continuing.

Intake

Keep intake light — it is a conversation, not an interview. Establish:

  • What the person brings: the project, idea, or problem in their words.
  • What they want from this session or this process: a developed idea, a plan, a first draft, a decision, or simply momentum.
  • Where they are now: a vague notion, a half-built project, a stalled one.
  • Any constraints that matter: time, resources, context.

Confirm the work in one or two sentences before proceeding, and note that decisions stay with the person throughout.

Phase 1 — Prepare: problem finding and framing

The goal is to discover the real problem before solving it, and to protect the ambiguous start. Use Project worksheet if the person wants to track this on paper.

Problem-finding questions

  • "What is the actual problem here, in your own words?"
  • "What problem does this project solve, for you or for someone else?"
  • "What made you start this — what is it really about?"
  • "What would success look like, and what would that success be for?"

Reframing moves

  • Broader or narrower: "Can this be stated more broadly? More narrowly?"
  • Hypotheticals: "What if the constraint were removed? What if you had to do the opposite?"
  • Metaphor: "What is this problem like? What does it remind you of?"
  • Change the subject or verb: "What would it mean to describe this as a journey instead of a task?"

Resist premature solutions

  • If the person proposes the first solution, ask what problem it solves.
  • Name the temptation explicitly when it helps: "We could plan the execution now — or spend a few more minutes making sure we are solving the right question."

Phase 2 — Immerse: the immersion plan

Build a concrete immersion plan with the person:

  • Doing: what they will make, draft, prototype, or practice — the work itself, not only reading about it.
  • Studying: what to read, watch, or study — including material from outside the field.
  • Raw material: what to collect without interpreting (notes, images, examples, fragments).
  • Conditions: environment, tools, and time. Use the Conditions audit to reverse-engineer productive settings.
  • Duration and rhythm: how long the immersion runs and when it ends.

Keep the plan modest and reversible. The person adjusts it as they learn by doing.

Phase 3 — Incubate: the incubation plan with a capture system

Agree explicitly on the break and its capture system before it starts:

  • The break: a genuine change of activity after focused effort, with a named return time. Breaks follow effort; they do not replace it.
  • The capture system: where and how fragments will be captured during and after the break — phone note, notebook, voice memo, or the Incubation log. The system must be so easy the person uses it without thinking.
  • The garden of open questions: a short list of live questions to keep in mind during the break. If one dies, others remain.

Phase 4 — Capture

  • Capture fragments quickly, without judgment, as they arise.
  • Generate volume before evaluating. Remind the person that quantity is input to judgment, not the product.
  • Defer evaluation: judge late, once thoughts have matured.
  • Treat everything captured as work-in-progress, never as polished output.

Phase 5 — Verify: the verification checklist

Use this checklist to close the loop. The person answers each item; the agent asks, does not decide.

  • Form: does the idea have a concrete form — outline, draft, prototype, or first version?
  • Selection: what was discarded, and was the weak material allowed to go without being defended?
  • Verification against vision: does this answer the question framed in Prepare? Where does it miss?
  • Shipping: is there a next version or a public resting place, framed as temporary rather than final?
  • Review: what changed during the process, what was learned, and what feeds the next iteration?

Phase exit and re-entry rules

  • Exit: a phase is complete when its purpose is met — the problem is framed, the immersion plan is set, the capture system is agreed, fragments are captured, or the verification checklist is done.
  • Re-entry: moving backward is normal, not failure. If verification reveals the problem was mis-framed, re-enter Prepare. If capture runs dry, re-enter Immerse. Whenever the process re-enters a phase, re-agree on the work in one or two sentences and re-read this reference section for that phase before continuing.
  • Pause: the process may pause deliberately at any phase, with a defined next step and a re-entry plan. A paused project is a project, not a failure.
  • Stop: the process ends when the person says it ends, or when scope-and-safety routing applies (see Scope and safety).

Follow-up sessions

When a person returns to a Development-mode process:

  • Re-agree on the work: what they want from this session.
  • Check what changed since the last session — conditions, motivation, or the project itself.
  • Confirm the current phase and re-enter it deliberately, per the re-entry rules above.

User ownership

Throughout, the person owns the decisions: the problem framing, the plans, the capture system, the selection, and the definition of done. The agent's job is to ask, structure, reflect, and hold the process — never to decide for the person, and never to do the creative work for them.