Skip to content
cccskills
FreePublished on GitHub

design

Explore competing approaches through independent agents before selecting an implementation.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md2.6 KB
  • agents/openai.yaml43 B

SKILL.md (original)

Review the instructions the agent will receive before you install.

Explore $ARGUMENTS and produce approaches with distinct core mechanisms. Use the smallest useful set of independent explorers; add one only to cover a material alternative or unresolved gap. If delegation is unavailable, disclose that limit and explore sequentially without claiming independence.

Establish the raw problem, actual constraints, and obvious solutions. Reframe where it exposes a different mechanism; do not force reframing for a bounded problem. Give explorers the same relevant repository context, their chosen method, and a distinct question or provocation. They must not see each other's proposals. Reframings are optional starting points, not constraints. Send revisions back to their authors.

Choose methods that expose different mechanisms; this is a menu, not required coverage:

MethodApproach and check
Build upwardStart with the smallest useful unit; check emergent needs and necessary upfront investment.
Work backwardDerive prerequisites from the ideal outcome; check the path is buildable.
Constraint firstSolve or dissolve the hardest constraints; preserve the user outcome.
Transfer an analogyBorrow a mechanism from another domain; identify where the analogy breaks.
EliminateRemove parts while preserving value; account for interactions and useful polish.
AdversarialDesign around consequential failures; state the cost of robustness.
ComposeCombine proven primitives; account for glue and dependency assumptions.
Evolve over timeShow the initial design, observable change triggers, and migration path; avoid speculative infrastructure.
Stakeholder trade-offsExpose conflicting needs and selectable compromises; prioritize rather than satisfy everyone equally.

Each explorer returns its mechanism, benefits, sacrifices, and conditions for choosing it. Compare mechanisms, merge cosmetic variants, and identify shared assumptions. Reopen exploration only when convergence or an uncovered assumption hides a useful option.

Save docs/designs/<short-name>.md with the problem and useful reframings, a landscape of approaches, their trade-offs and selection conditions, shared assumptions, and consequential unexplored territory. Include diagrams when they explain architectural or spatial choices. Stop after presenting the alternatives; implementation requires the user's selection or authorization.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

audit

Free

Review code through focused dimensions, with one independent agent per dimension.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

Distill rough text into dense, expert-register English where terms of art carry the intent.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

Apply personal Git conventions when committing, creating or updating PRs, merging, rebasing, or explicitly shipping work end to end.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

judge

Free

Obtain an independent review of the approach and trade-offs in completed work.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

Refine a supplied question into a precise, expert-level query when the user asks to rewrite or optimize their question or prompt. Do not use to answer the underlying question.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

sota

Free

Unanchored expert verdict on the open decision, checked against the current state of the art.

The English summary is being prepared. Showing the original description.

JHostalek/dotclaude11Updated Oct 16, 2026

See all skills by JHostalek

Report a problem with this skill