By phyr97
Structured ideation panels with dynamic personas, critic, and synthesizer. Run /ideate on any open decision question.
Step-Back pre-phase agent. Abstracts the problem to higher-level concepts and first principles before Round 1 starts. Output is injected into the proposers' context.
Devil's advocate in the expert panel. Reads all proposer outputs and produces a structured list of weaknesses, hidden assumptions, and failure modes. Uses Steelman-Solitaire by default, or Chain-of-Verification when --verify is passed.
Evolution agent in the expert panel. Given the top-2 proposals from the synthesis, mutates them by combining elements from other proposals or fixing individual weaknesses.
Proposer in the expert panel. Generates ideas from a specific persona's perspective. Spawned once per persona per round; receives persona definition and problem as spawn prompt.
Rough implementation plan agent in the expert panel. Given the final recommendation, produces 3-8 high-level steps with rough timing and dependencies. Not a detailed implementation plan.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that turns any open decision or ideation question into a defensible recommendation. Runs a structured multi-agent panel: dynamic personas, a devil's advocate, and a synthesizer that produces a top pick with explicit trade-offs.
A single-shot LLM answer is a single-perspective answer. Panels with heterogeneous proposers, a dedicated critic, and an explicit synthesizer produce clearer recommendations, especially when the problem is unclear or has competing constraints.
/plugin marketplace add phyr97/phyr97-marketplace
/plugin install expert-panel@phyr97
claude --plugin-dir /path/to/expert-panel
Use /reload-plugins inside a session after edits.
Subagents may need explicit tool permissions. Add to ~/.claude/settings.json:
{
"permissions": {
"allow": ["WebSearch", "WebFetch", "Read", "Grep"]
}
}
This lets the critic verify factual claims when --verify is active.
/ideate "should we use Redis or Memcached as session store"
Runs the default standard mode: 4 proposers, 2 rounds, critic, synthesizer.
| Mode | Activates | Relative cost |
|---|---|---|
--mode quick | 3 proposers, 1 round | ~0.7x |
--mode standard (default) | 4 proposers, 2 rounds | 1.0x |
--mode thorough | + step-back, verify, monitor-diversity | ~1.6x |
--mode max | + elo, evolve | ~3-4x |
| Flag | Effect |
|---|---|
--step-back | Abstraction pre-phase before Round 1 |
--interactive | Pause after Round 1 for user direction |
--verify | Critic verifies factual claims with tool use |
--monitor-diversity | Checks persona outputs for redundancy, retries if too similar |
--elo | Pairwise tournament synthesis instead of one-shot |
--evolve | Mutate the top proposals with cross-pollination |
--plan | Add a rough implementation plan after synthesis |
--debate | Teams mode instead of subagents (experimental) |
| Flag | Default | Range | Effect |
|---|---|---|---|
--rounds N | 2 | 1-3 | Max rounds of proposer iteration |
--agents N | 4 | 3-5 | Number of proposer personas |
--max-tokens N | none | any | Best-effort token cap |
--estimate | off | flag | Dry-run with token breakdown |
Terminal: top recommendation, 2-3 alternatives with their conditions, top 3 risks, plus the path to the full report.
Markdown report at ~/.claude/ideate/YYYY-MM-DD-HHmm-<slug>.md: full synthesis, persona details, aggregated arguments, critique highlights, optional sections (plan, evolution, elo).
Metrics at ~/.claude/ideate/metrics.jsonl: one JSON line per run for later analysis.
Single orchestrator skill plus a set of subagents:
ep-persona: proposer (one per persona per round)ep-critic: devil's advocateep-synthesizer: recommendation plus alternativesep-abstracter: optional, step-back phaseep-evolver: optional, mutates top proposalsep-planner: optional, rough planCLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 is not set.MIT. See LICENSE.
Modular deep research plugin for Claude Code with auto-scaling, capped adversarial verification, and self-check
Analyzes how well Claude Code plugins perform by evaluating JSONL session transcripts
Multi-mode code review with agent-based analysis, false-positive filtering, and confidence scoring
npx claudepluginhub phyr97/phyr97-marketplace --plugin expert-panelMulti-perspective council analysis. Spawns parallel cognitive perspectives to analyze questions, plans, and ideas from multiple angles with structured dialectical synthesis.
Expert panel discussion system with 22 curated personas. Includes 2 agents, 1 command, and 22 expert definitions.
Convene parallel role-specialized peer agents (dynamic roster, plan card first) to debate a cross-domain decision or audit a codebase in real time. Invoking Claude acts as CEO: convenes, routes peer-DMs, arbitrates deadlocks, writes a one-page decision log.
Structured thinking methods that counteract LLM reasoning biases during problem exploration — first principles, inversion, constraint manipulation, perspective forcing, analogy search, and more, with user-gated parallel subagent exploration for deep dives
Structured thinking methods as multi-agent sessions — Six Thinking Hats, Adversarial Debate, Disney Method, Pre-Mortem, Polarity Management, TRIZ, W³, Troika Consulting, Wise Crowds, First Principles Decomposition, Analogical Transfer, Morphological Box, Futures Cone — plus meta-cognitive modules (Bias Check, Frame Check, Method Check) that reflect on the thinking itself
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.