agentsclimarketplace

Design

Skill ronniepinnell/casper/collection/research-and-analysis/design

Interactive design session. Agent team debates architecture, captures decisions, logs ideas.From its SKILL.md

Install
npx -y skills add ronniepinnell/casper --skill design

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, 445 tokens by cl100k_base, as published. Nobody here has run it

/design — Interactive Design Session

Enters DESIGN mode. Operator drives the conversation, Claude facilitates with domain experts.

What Happens

  1. Load context: Read {plan_file} "Current State", relevant specs, load recent decisions via storage list_decisions
  2. Identify domain: Match topic against .claude/agents/AGENTS_GUIDE.md routing table
  3. Facilitate debate: Present trade-offs, ask Operator for decisions, challenge assumptions
  4. Capture continuously: Decisions → storage record_decision, Ideas → storage record_idea
  5. On /done: Run content pipeline (see .agents/reference/knowledge_capture.md)

Team Mode (if CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS is set)

Lead + Spec Writer + Idea Logger + Domain Expert(s) from routing table. Teammates debate each other, not just report to lead.

Fallback Mode (if env var not set)

Sequential subagent calls:

  1. Domain expert agent(s) for research/analysis
  2. Main context facilitates Operator discussion
  3. Spec writer agent for doc updates at end

Mode Rules

  • Explicit entry only — Operator invokes /design
  • Nests: /meeting or /quick can run inside /design
  • /done pops back to outer mode (or exits if no outer)
  • /done session clears entire stack + runs close-out

Gate References

  • Gate 1 (pre-push): Ensures design artifacts sync to specs
  • Gate 2 (pre-commit): Validates prompt quality if prompts generated

Judgment weave (see /judgment)

  • Diverge phase: /think moves are the debate fuel — assign different agents different moves (one runs invert, one base-rate, one analogy) instead of N agents free-associating.
  • Converge phase: every surviving option through /door; one-way doors get the five questions in the room.
  • Exit: chosen design gets /premortem + /gate lines; all decisions → /verdict log (in addition to any project decision store).

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,569. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.