PaperclipOrg
Claude Agent Skill · by Sickn33

Multi Agent Brainstorming

Install Multi Agent Brainstorming skill for Claude Code from sickn33/antigravity-awesome-skills.

Install
Terminal · npx
$npx skills add https://github.com/obra/superpowers --skill brainstorming
Works with Paperclip

How Multi Agent Brainstorming fits into a Paperclip company.

Multi Agent Brainstorming drops into any Paperclip agent that handles this kind of work. Assign it to a specialist inside a pre-configured PaperclipOrg company and the skill becomes available on every heartbeat — no prompt engineering, no tool wiring.

S
SaaS FactoryPaired

Pre-configured AI company — 18 agents, 18 skills, one-time purchase.

$27$59
Explore pack
Source file
SKILL.md261 lines
Expand
---name: multi-agent-brainstormingdescription: "Simulate a structured peer-review process using multiple specialized agents to validate designs, surface hidden assumptions, and identify failure modes before implementation."risk: unknownsource: communitydate_added: "2026-02-27"--- # Multi-Agent Brainstorming (Structured Design Review) ## Purpose Transform a single-agent design into a **robust, review-validated design**by simulating a formal peer-review process using multiple constrained agents. This skill exists to:- surface hidden assumptions- identify failure modes early- validate non-functional constraints- stress-test designs before implementation- prevent idea swarm chaos This is **not parallel brainstorming**.It is **sequential design review with enforced roles**. --- ## Operating Model - One agent designs.- Other agents review.- No agent may exceed its mandate.- Creativity is centralized; critique is distributed.- Decisions are explicit and logged. The process is **gated** and **terminates by design**. --- ## Agent Roles (Non-Negotiable) Each agent operates under a **hard scope limit**. ### 1️⃣ Primary Designer (Lead Agent) **Role:**- Owns the design- Runs the standard `brainstorming` skill- Maintains the Decision Log **May:**- Ask clarification questions- Propose designs and alternatives- Revise designs based on feedback **May NOT:**- Self-approve the final design- Ignore reviewer objections- Invent requirements post-lock --- ### 2️⃣ Skeptic / Challenger Agent **Role:**- Assume the design will fail- Identify weaknesses and risks **May:**- Question assumptions- Identify edge cases- Highlight ambiguity or overconfidence- Flag YAGNI violations **May NOT:**- Propose new features- Redesign the system- Offer alternative architectures Prompting guidance:> “Assume this design fails in production. Why?” --- ### 3️⃣ Constraint Guardian Agent **Role:**- Enforce non-functional and real-world constraints Focus areas:- performance- scalability- reliability- security & privacy- maintainability- operational cost **May:**- Reject designs that violate constraints- Request clarification of limits **May NOT:**- Debate product goals- Suggest feature changes- Optimize beyond stated requirements --- ### 4️⃣ User Advocate Agent **Role:**- Represent the end user Focus areas:- cognitive load- usability- clarity of flows- error handling from user perspective- mismatch between intent and experience **May:**- Identify confusing or misleading aspects- Flag poor defaults or unclear behavior **May NOT:**- Redesign architecture- Add features- Override stated user goals --- ### 5️⃣ Integrator / Arbiter Agent **Role:**- Resolve conflicts- Finalize decisions- Enforce exit criteria **May:**- Accept or reject objections- Require design revisions- Declare the design complete **May NOT:**- Invent new ideas- Add requirements- Reopen locked decisions without cause --- ## The Process ### Phase 1 — Single-Agent Design 1. Primary Designer runs the **standard `brainstorming` skill**2. Understanding Lock is completed and confirmed3. Initial design is produced4. Decision Log is started No other agents participate yet. --- ### Phase 2 — Structured Review Loop Agents are invoked **one at a time**, in the following order: 1. Skeptic / Challenger2. Constraint Guardian3. User Advocate For each reviewer:- Feedback must be explicit and scoped- Objections must reference assumptions or decisions- No new features may be introduced Primary Designer must:- Respond to each objection- Revise the design if required- Update the Decision Log --- ### Phase 3 — Integration & Arbitration The Integrator / Arbiter reviews:- the final design- the Decision Log- unresolved objections The Arbiter must explicitly decide:- which objections are accepted- which are rejected (with rationale) --- ## Decision Log (Mandatory Artifact) The Decision Log must record: - Decision made- Alternatives considered- Objections raised- Resolution and rationale No design is considered valid without a completed log. --- ## Exit Criteria (Hard Stop) You may exit multi-agent brainstorming **only when all are true**: - Understanding Lock was completed- All reviewer agents have been invoked- All objections are resolved or explicitly rejected- Decision Log is complete- Arbiter has declared the design acceptable- If any criterion is unmet:- Continue review- Do NOT proceed to implementationIf this skill was invoked by a routing or orchestration layer, you MUST report the final disposition explicitly as one of: APPROVED, REVISE, or REJECT, with a brief rationale.--- ## Failure Modes This Skill Prevents - Idea swarm chaos- Hallucinated consensus- Overconfident single-agent designs- Hidden assumptions- Premature implementation- Endless debate --- ## Key Principles - One designer, many reviewers- Creativity is centralized- Critique is constrained- Decisions are explicit- Process must terminate --- ## Final Reminder This skill exists to answer one question with confidence: > “If this design fails, did we do everything reasonable to catch it early?” If the answer is unclear, **do not exit this skill**. ## When to UseThis skill is applicable to execute the workflow or actions described in the overview. ## Limitations- Use this skill only when the task clearly matches the scope described above.- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.