When a decision needs
a second look.
RelayLaunch prepares recovery actions, workflow recommendations, and owner approvals. Specialist Review is the safety layer for the decisions that deserve more scrutiny: different perspectives, dissent tracking, and an audit trail before anything important moves forward.
Examples this layer can review
Not one AI. A council.
Each archetype brings a fundamentally different lens. Strategic rivalry. Analytical precision. Historical pattern-matching. The insight lives in the disagreement.
Olympus Council
Fourteen directors who compete as much as they collaborate. Strategic rivalry surfaces options that groupthink buries.
Strategy Room
Observe, Orient, Decide, Act. Five analysts trained for decisions measured in hours, not quarters.
Founders Board
Five founders who've built before. They recognize the patterns because they've lived them, and survived them.
Titans Boardroom
Fourteen business titans. They think in market share, competitive moats, and quarterly earnings. Not vibes.
The Station
Fourteen non-human intelligences. When every human council agrees, The Station asks the question nobody thought to ask.
Elements
Five primal forces: fire, water, earth, wind, void. They strip decisions down to first principles before building back up.
Seventeen council modes. Seventeen ways of thinking.
Each council is a different universe of expertise. Hover to feel the shift. Click to step inside.
Olympus Council
Strategic tension and rivalry
14 Olympian directors with built-in rivalries. Ares pushes revenue, Athena demands strategy, Hermes wants speed. Strategic tension produces better decisions.
Inside Olympus
- Ares Sales — pushes aggressive growth
- Athena Operations — demands sustainability
- Hermes Growth — finds the fastest path
- Apollo Brand — protects the story
- Hephaestus Build — architects the system
- Artemis Intelligence — hunts blind spots
Founders Board
Historical business pattern-matching
First-principles reasoning, contrarian bets, operational discipline. Seven advisors modeled on history's greatest entrepreneurs.
Inside Founders Board
- First-Principles Thinker Strips every assumption bare
- The Operator Runs unit economics cold
- The Contrarian Bets against consensus
- Market Timer Reads cycles and inflection points
- Scaling Architect Maps proven growth patterns
- The Capital Architect Structures funding, runway, and burn rate
- The Culture Builder Designs teams, org structure, and hiring strategy
Strategy Room
Focused precision and strategic analysis
OODA loop: observe, orient, decide, act. Three courses of action. Red Cell stress-tests each one. The Founder decides.
Inside Strategy Room
- Research Analyst Gathers facts on the ground
- Strategist Builds 3 Courses of Action
- Red Cell Stress-tests every plan to break it
- Logistics Lead Scores feasibility and cost
- Decision Lead Makes the final call
- Comms Lead Manages stakeholder messaging during crisis
- Rear Guard Protects working systems while fighting the fire
Based on OODA loop and structured decision analysis.
The Agora
Customer empathy and market voice
Named after the ancient Greek marketplace. Seven audience archetypes react to your idea, pricing, or messaging before real customers ever see it.
Inside Agora
- The Patron Would I buy this?
- The Skeptic What's the catch?
- The Ambassador Would I tell a friend?
- The Newcomer Am I confused?
- The Merchant Is the price fair?
- The Churner Why did I leave? What would make me cancel?
- The Power User Deep engagement demands and advanced needs
Hear the market before you commit.
Dark Forest
Game theory and adversarial thinking
What if your competitor is already planning what you're planning? Every market move reveals your position. Game theory applied to business strategy.
Inside Dark Forest
- Hunter Maps competitor positions and intent
- Cloaker Conceals your strategic moves
- Signal Analyst Reads what market moves reveal
- Strike Planner Times offensive positioning
- Deterrence Architect Builds defensive moats
- The Cartographer Maps the full competitive landscape and alliances
- The Double Agent What if our own strategy is being used against us?
Based on Liu Cixin's Dark Forest theory.
The Lab
Evidence-based hypothesis testing
Form a hypothesis. Design the experiment. Gather evidence. What's the cheapest test that would change the decision?
Inside Lab
- Hypothesis Architect Frames the testable question
- Experiment Designer Builds the cheapest valid test
- Evidence Gatherer Collects data, not opinions
- Statistical Reviewer Validates significance
- Peer Review Panel Challenges every conclusion
- The Replicator Validates findings across different conditions
- The Falsifier Actively tries to disprove every result
Future Council
Trend extrapolation and foresight
30/90/180-day forecasts with Bayesian confidence. Identifies inflection points where small actions create large divergences.
Inside Future
- Trend Analyst Reads signals others miss
- Bayesian Forecaster Assigns real probabilities
- Black Swan Hunter Watches for what nobody expects
- Scenario Planner Builds 3 divergent futures
- Precog Self-calibrating confidence scores
- The Historian Past patterns that predict future moves
- The Accelerationist What happens faster than anyone expects?
The Chronicle
Narrative architecture and storytelling
Seven masters of language shape your message. The Storyteller builds the arc, The Editor cuts the fluff, The Bard finds the emotional hook, The Translator adapts per audience.
Inside Chronicle
- The Storyteller Builds the narrative arc
- The Editor Cuts to the one sentence
- The Bard Finds the emotional hook
- The Scribe Guards brand consistency
- The Muse Sparks the unexpected angle
- The Translator Adapts the message per audience segment
- The Archivist Institutional memory — what's been tried before
Daemon Garden
Background insight cultivation
Seven autonomous daemons tend seeds of insight between sessions. Insights compound while you sleep.
Inside Daemon Garden
- Seed Planter Spots early signals before they bloom
- Root Tender Connects distant, unrelated ideas
- Mycelium Runner Moves insights between sessions
- Bloom Watcher Knows when insight is ready to harvest
- Compost Breaker Recycles failed strategies into fuel
- Canopy Keeper Maintains the big picture
- Pollinator Cross-fertilizes between your projects
MetaClaw
Recursive self-examination
The system thinking about itself. Audits agent performance, finds blind spots, and proposes its own upgrades.
Inside MetaClaw
- Logic Auditor Scores reasoning quality of every output
- Adversary Stress-tests conclusions with hostile prompts
- Blind Spot Scanner Finds what other councils missed
- Upgrade Architect Proposes system-level improvements
- Rule Rewriter The only council that can change itself
- The Calibrator Measures confidence vs actual accuracy
- The Genealogist Traces decision lineage and inherited biases
Symbiosis
Human-AI co-evolution
Models your thinking patterns and cognitive blind spots. Adapts the system to amplify your strengths.
Inside Symbiosis
- Behavioral Modeler Tracks your decision patterns
- Cognitive Mirror Reflects your biases back at you
- Amplifier Strengthens your natural advantages
- Compensator Fills the gaps you can't see
- Evolution Tracker Measures how your decisions improve
- The Integrator Combines human and AI workflows into one system
- The Divergent Knows when to break the pattern, not follow it
The Tribunal
Justice, accountability and governance
Seven seats of deliberation examine every decision for who it serves, who it harms, and whether you can defend it in two years.
Inside Tribunal
- The Advocate Who's not being heard?
- The Judge Weighs both sides fairly
- The Prosecutor What could go wrong?
- The Witness What do we actually know?
- The Steward Will we be proud of this?
- The Precedent What happened last time? Historical case analysis
- The Oracle Downstream consequences at 1, 5, and 10 years
Titans Boardroom
Executive power plays and market dominance
Fourteen business titans who think in market share, competitive moats, and quarterly earnings. Executive power plays surface the moves that polite strategy decks bury.
Inside Titans Boardroom
- The Dealmaker Structures growth partnerships
- The Operator Runs P&L with surgical precision
- The Lobbyist Navigates regulation and policy
- Market Maker Reads capital flow and liquidity
- The Strategist Builds competitive moats
- The Rainmaker Closes revenue at scale
- The Fixer Solves the unsolvable political problems
The Station
Non-human radical perspectives
Fourteen non-human intelligences. When every human council agrees, The Station asks the question nobody thought to ask. Radical perspectives from outside the species.
Inside Station
- Xenoanalyst Frames problems outside human bias
- Pattern Breaker Finds assumptions baked into every question
- Deep Time Thinks in centuries, not quarters
- Emergence Spots complexity no one designed
- The Outsider Asks 'why do humans assume this?'
- Signal Ghost Detects what's absent from the data
- Nexus Mind Synthesizes across all non-human perspectives
Elements
Elemental first-principles reasoning
Seven primal forces — fire, water, earth, wind, void, light, shadow. They strip decisions down to first principles before building back up. Elemental balance reveals what overcomplicated analysis obscures.
Inside Elements
- Ignis (Fire) Intensity, drive, creative destruction
- Aqua (Water) Adaptability, flow, finding the path of least resistance
- Terra (Earth) Stability, structure, what's been proven
- Ventus (Wind) Speed, change, communication
- Aether (Void) The space between — what's missing from the equation
- Lux (Light) Clarity, transparency, what's being hidden
- Umbra (Shadow) Hidden costs, unspoken trade-offs, the unseen
See what the first answer may miss
Every review mode brings a different lens. The recommendation matters, but the dissent is where risk, timing, and tradeoffs become visible.
“Should I go back to school at 35 or stay in my career?”
Go back, but not full-time. A part-time program keeps income flowing while building the credential. The real risk isn't tuition, it's identity drift: you'll question the decision every hard week. Pick a program with a cohort so you have peers in the same boat. Your career experience is an asset in the classroom, not a liability.
- Apply to 2-3 part-time programs this cycle
- Calculate your runway at 60% income for 18 months
- Talk to 3 people who made this switch after 30
“Production database 5x error rate, customers complaining”
Targeted query kill, not full rollback. The spike correlates with the 2:14 PM release that introduced an unindexed JOIN on the orders table. Kill that query class, add the index, and ship again. Full rollback risks losing 3 hours of clean transactions.
- Kill long-running queries matching the pattern NOW
- Add composite index on orders(customer_id, created_at)
- Redeploy with the fix, monitor for 30 minutes
- Post-incident review within 24 hours
“I have two completely different directions for my novel. Which one?”
Write the one that scares you. The 'safe' concept has a clear market but you'll lose steam at 40,000 words because you've seen it before. The weird one, the one you're not sure anyone else will get, that's the one with voice. Readers don't remember competent. They remember singular. Outline both for one week, then commit to whichever one you're still thinking about at 2 AM.
- Outline both concepts for 5 days, no commitment yet
- Write the opening scene of each and read them aloud
- Ask: which one would you regret NOT writing?
Describe the decision. Pick a review mode. Approve with context.
Use this for recovery actions, workflow changes, pricing questions, hiring decisions, or any recommendation that should be challenged before it moves.
What decision is your firm facing?
Describe the situation with enough detail for a board-level debate.
Review a real business action
Bring an action you are considering. RelayLaunch will show the recommendation, the objections, and the next step to approve.
3 free analyses per session. No signup required.
Watch the review unfold.
Select specialist perspectives, ask your question, and see the recommendation challenged in real time. Dissent, opposing views, and visible reasoning keep the owner in control.
Consensus is easy. Disagreement is where the value lives.
Every AI tool gives you an answer. RelayLaunch gives you the review behind the answer, and every review has an audit trail. The Dissent Score tells you how much the specialists disagree, and high dissent means the owner should slow down before approving.
Dissent Score
High Dissent0%
Dissent
3 of 5 councils disagree on the recommended approach. The minority sees risks the majority is discounting.
Disagreement Ratio
3 of 5 councils dissent
Insight Value
Valuable - multiple perspectives clash
Dissent Evolution
↑ 48ptsOpposing Views
Every dissenting voice is recorded. Not hidden. Displayed first. The minority sees what the majority misses.
Disagreement Score
A 0-100 gauge measuring how much experts disagree. High disagreement = high insight value. Low disagreement = consensus (less interesting).
"Why This?" Links
Every output explains itself. Every score has a reason. Every dissent cites evidence. No black boxes.
The Founder's Strategy Room
Your AI strategy cabinet. Honest counsel, not yes-men. Daily briefings, dissent alerts, blind spot detection, and a decision journal that tracks when the minority was right.
Daily Briefing
Preview: Your Daily Briefing
High-Dissent Alerts
2
Pricing strategy: should we raise or hold?
4/5 teams disagree
Hiring timeline: can we delay Q3 hires?
3/5 teams disagree
Decision Journal
Decisions this month
12
Minority was right
3 (25%)
Avg dissent score
58
Blind Spots
“You haven't stress-tested your go-to-market assumptions in 14 days”
14 days since last checked
New Council Modes. New milestones. On the roadmap.
Consensus Heatmap
Visualize where review modes agree and disagree on any question. Discover the outlier perspectives that add the most value before an action is approved.
The Dreamscape
A verse where subconscious patterns surface. Lateral leaps, unexpected connections, and the ideas your rational mind filters out. Sometimes the best strategy comes from a dream.
The Underworld
Shadow analysis: the hidden costs, buried assumptions, and uncomfortable truths no one wants to say out loud. This verse digs where others won't look.
The Reckoning
A milestone review that revisits your most contentious decisions with historical context. One question, multiple perspectives, and a clear owner approval record.
Shareable Review Records
The Chronicle records every analysis. Share recommendation links with your team, investors, or audience: the full review, disagreement timeline, and opposing views. Auditable intelligence you can point to.
Try the Playground → Share a RecommendationThis layer challenges recommendations before they reach you.
Owner-approved operations need more than a confident draft. RelayLaunch uses review guardrails that make agreement harder, not easier, before a high-impact action moves forward.
No Premature Agreement
Specialists cannot agree without citing evidence anchors. No evidence, no approval-ready recommendation.
Mandatory Disagreement Threshold
If disagreement drops too low on a risky action, an opposing review perspective is assigned. Groupthink triggers intervention.
Conviction Anchoring
Positions are weighted by evidence quality: empirical (1.0x), synthetic (0.7x), inferred (0.5x), unsupported (0.1x).
Anti-Capitulation
Once a specialist takes a position, it cannot reverse without introducing new evidence. No flip-flopping under pressure.
Recovery Engine prepares the next action. Specialist Review challenges the recommendation. Relay Deck keeps the owner in control with visible reasoning, dissent, and proof records. The goal is simple: useful automation without hidden autopilot.
Discoverable by every enterprise AI platform.
RelayLaunch's review layer is built for interoperable agent workflows. Protocol surfaces like A2A and MCP make it possible for other systems to discover review capabilities without custom one-off integrations.
A2A Protocol v2.2
Google's Agent-to-Agent protocol. Agent Card at /.well-known/agent.json with 18 discoverable skills. JSON-RPC 2.0 bridge at /api/a2a.
MCP Server
Model Context Protocol for tool-based AI access. 4 tools: invoke_council, list_modes, get_verdict, council_status.
CaaS REST API
Specialist review as an API. One POST, multiple AI analyst perspectives, with signed response records.
POST /api/v1/caas/debate npm Package
Open-source Council Modes. Use portable review structures with any AI model provider.
npm i @relaylaunch/councilverse-formations Embed Widget
Embed any recommendation on your site with one line. Branded card with confidence + disagreements.
<iframe src="deck.relaylaunch.com/embed/verdict/..."> n8n Node
Trigger AI reviews from any n8n workflow. Use auto-select to route the right review mode.
npm i n8n-nodes-CouncilVerse Webhooks
Signed verdict.created events pushed to your endpoints in real time.
SHA-256 HMAC · Retry with backoff Every analysis makes the next one smarter.
Three compounding loops run as the system collects approved work, rejected work, and review history. Your decision intelligence deepens as real usage accumulates.
Past decisions become instant references.
Semantic search over your decision history. Similar questions surface relevant precedents so new analyses start with institutional knowledge, not a blank slate.
The system tracks which council modes work best.
Continuous quality scoring across all analyses. Auto-recommends the optimal council mode for each question type. Your specialists sharpen with every use.
Disagreement becomes strategic foresight.
When models disagree, the system catalogs why. Future analyses surface historical disagreement patterns, turning opposing views into early warning signals.
Digest surfacing the most significant minority opinions from your reviews. See what the first answer missed before it matters.
Interactive visualization of the available review modes. See how specialists connect and vote. Explore methodology structures before launching an analysis.
Build your own councils
The Council Modes and voting engine are MIT-licensed. Use them with any AI model provider. The full platform adds persistence, audit trails, and compliance.
# Scaffold a project in 60 seconds
npx create-councilverse my-council
# Or install the packages directly
npm install @relaylaunch/councilverse-formations
npm install @relaylaunch/councilverse-votingStart free. Add more councils as you grow.
You bring your own AI key and pay the model provider directly. We charge for the platform, not the thinking.
Free
- Up to 4 councils per question
- 3 analyses per session
- Last 5 analyses saved
- 4 councils: Olympus, Founders, Strategy Room, The Agora
Pro
- All Council Modes
- Unlimited analyses
- Full analysis history
- AI Chief of Staff remembers your context
- Export analyses as PDF
Team
- 5 team seats included
- All Council Modes + priority queue
- Shared workspace + analytics
- AI Chief of Staff deep memory + adaptation
- API access included
Enterprise
- Unlimited team seats + SSO
- All Council Modes + custom creation
- Full audit trails + compliance
- Org-level AI Chief of Staff memory and adaptation
- Dedicated support + on-premise option
🛡️ Try Free. Cancel Anytime.
3 free analyses per session, no signup required. All paid plans are month-to-month. Cancel anytime, no penalty.
Built for owner-reviewed decisions
Named specialists with distinct reasoning styles, specialties, and perspectives for the decisions that need a second look.
Strategy, risk, legal, market, design, operations, and founder review methods that route the right specialists to the right question.
Bring your own API key. You pay the model provider directly. We never store your credentials.
Ask a question, pick a review mode, and get a recommendation with visible reasoning quickly enough to use in the workday.
Meet Your AI Chief of Staff
Your Guide Between Worlds
Your AI Chief of Staff listens to your question, selects the right specialists, moderates the review, and synthesizes the recommendation. She routes the owner toward the next best approved action. Not a chatbot. A reasoning guide.
What the Platform Does
Early access. Dogfood and friendly pilot proof lanes in progress.
Every AI analysis can route across 7 live cloud providers and 17 live models. Different training data surfaces different risks. Disagreements are preserved, never hidden.
Every recovery recommendation waits for direct owner approval. Once confirmed and completed, it builds an exportable revenue ledger that you own and can share directly with your accountant.
AI specialist coverage across every business function. Marketing, sales follow-up, security, content, systems, and daily operations stay coordinated for you.
Average AI analysis costs ~$0.03 using optimized multi-model routing. DeepSeek V4-Flash runs at 21x less than Claude Sonnet for routine tasks.
Your data stays portable. Export recommendations, reasoning traces, and deliverables anytime. No lock-in. Cancel and take everything with you.
In early pilot testing, the Recovery Engine surfaced 40% fewer no-shows (wellness practice) and $7,800 in recovered revenue (auto repair, first 30 days). All metrics are projected from pilot program design.
Stop rubber-stamping AI answers. Review the action first.
Recovery Engine prepares the action. Specialist Review challenges it. The owner approves what moves.
🛡️ Proof-layer review · Month-to-month, cancel anytime · Try a sample before you commit