Most decisions fail for invisible reasons.
We make them visible.

Paste your decision. See what you're missing.

GET STARTED
Try it free

10 decisions per month. Start free.

HOW IT WORKS
Four steps

1. Describe the decision

One paragraph. That's it.

2. We break it down

Your reasoning. Hidden assumptions. Expected outcomes. Key risks.

3. You track reality

Compare what you expected with what actually happened.

4. You learn what was right

Not opinions. Actual feedback on your thinking.

BEFORE YOU GET ACCESS

Everything you need to know in 60 seconds

What is this?

OMEGA is a tool that helps you make better decisions. You describe a decision — in plain English — and it shows you the hidden assumptions, the risks, what you should check before committing, and what to track afterwards.

What is an API?

An API is just a way for one piece of software to talk to another. Think of it like a plug socket — you plug in, electricity comes out. You plug a decision in, structured thinking comes out. You don't need to understand how it works any more than you understand how electricity works.

What is the access code?

When you sign up, you get a unique code. It looks like this: om_live_abc123. This is your personal key. It identifies you and saves your decisions to your account. Think of it like a password — except you only need it once. After that, the tool remembers you.

Do I need to write code?

No. You can use OMEGA in three ways:

1. On this website

Scroll up. Type a decision. Click the button. That's it. No code, no setup, no installation. Works right now.

2. In your own app or tool

If you're a developer, you can connect OMEGA to your website, internal tool, or AI agent using a few lines of code. Every decision your users make gets traced automatically.

3. Through an AI agent

If you're building autonomous AI systems, plug OMEGA in so every action the agent takes is checked, traced, and governed — before it executes.

What do I get for free?

10 decisions per month. Each decision gives you: a breakdown of your reasoning, a list of hidden assumptions, the key risks, what to check before committing, and a permanent tamper-proof record. Come back any time to check if your predictions were right.

What happens after I sign up?

Step 1: Enter your invite code below. You get your access code instantly.

Step 2: Scroll to the decision box at the top of this page.

Step 3: Pick an example or type your own decision.

Step 4: Click 'Decode this decision'. See your results in seconds.

Step 5: Come back any time. Your decisions are saved. You don't need to enter the code again.

No software to install. No account to create. No credit card. Just a code and a decision.

Who is this for?

Anyone making a decision that matters. Founders deciding whether to expand. Scientists deciding which hypothesis to test. Teams deciding what to build next. Investors deciding where to put money. Compliance officers who need to prove decisions were made responsibly.

If the decision is important enough to get right, this is for you.

How do developers connect it to their own tools?

Send a decision to OMEGA and get structured results back. Here's all it takes:

// Send a decision
const response = await fetch('https://api.omegaprotocol.org/v1/assumptions/extract', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_ACCESS_CODE',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    text: 'Should we expand into the US market?',
    session_id: 'strategy_decision_01'
  })
});

// Get back: list of hidden assumptions, each one trackable
const result = await response.json();

That's it. One call. Works from any programming language, any platform, any tool. Full docs below ↓

FOR DEVELOPERS
APIs, endpoints & integration

Everything below is technical reference: seven APIs, architecture, live code examples, and full documentation. Base URL: https://api.omegaprotocol.org/v1

APPLICATIONS
AI agents · Autonomous systems · Scientific research · Regulated industries · Boards & founders
DECIDE API
Structured multi-lens analysis + governance state machine
ASSUMPTIONS API
Extract · Track · Test · Invalidate premises
TRACE API
Hash-chained records · Replay · Verification
EXPECTATION API
Baseline · Surprise · Calibration
CONFIRMATION API
Intent · Gate · Non-action as data
OMEGA PROTOCOL
Open trace + governance data model · MIT licensed
DECIDE automatically generates trace records, assumption registries, and compliance evidence — so every decision is auditable, reproducible, and governed before it ships. Set expectations before acting, record confirmation gates, and score surprise when outcomes resolve.
Use DECIDE end-to-end, or call any of the seven APIs independently.
PLATFORM
Seven APIs. One session model.
Use each API independently, or use DECIDE to get all seven in one call. Every API shares the same session, auth, and trace chain.
◈

Trace

Hash-chained, tamper-evident records for every action and decision. Replay any chain. Verify integrity. Export for audit. Real-time stream and compliance reports.

POST /trace/record
GET /trace/chain/{session}
POST /trace/verify
POST /trace/replay
GET /trace/export/{session}
GET /trace/compliance-report/{sessionId}?framework=eu_ai_act
GET /trace/stream/{sessionId} (SSE)
GET /dashboard/{sessionId}
△

Assumptions

Extract assumptions from any text. Track status over time. Test against new evidence. Surface what's unvalidated and risky.

POST /assumptions/extract
PATCH /assumptions/{id}
POST /assumptions/test
GET /assumptions/registry/{session}
GET /assumptions/risk
◆

Decide

Multi-lens decision analysis with governance. Six analytical modes, assumption extraction, tension detection, and a decision state machine. Automatically creates traces and extracts assumptions.

POST /decide/analyze
POST /decide/stress
POST /decide/rerun
GET /decide/{session}/tensions
GET /decide/{session}/summary
◉

Expectation

Record what was expected before a decision. Set a baseline. Score surprise when reality arrives.

POST /expectation/set
POST /surprise/score
✓

Confirmation

Separate intent from action. Record provisional decisions and whether they were confirmed or abandoned before execution.

POST /confirmation/record
∿

Surprise

Score how far reality diverged from expectation. Surface which assumptions were invalidated by the outcome.

POST /surprise/score
◇

Agent

Agent Guard: pre-execution checks for autonomous agents. Boundary enforcement, risk assessment, and escalation gates.

POST /agent/check
ARCHITECTURE
One session, seven artifact types
All seven APIs attach to the same session. Query any session to get the full picture.
// Every session contains all seven artifact types
{
  "session_id": "ses_trial_2026_03",
  "trace_records": [12 records, hash-chained],
  "assumptions": [8 extracted, 3 tested, 1 invalid],
  "decisions": [2 analyses, state: "CONDITIONAL"],
  "expectations": [3 baselines set, 1 resolved],
  "confirmations": [4 recorded, 2 acted, 2 no_act],
  "agent_checks": [6 proposed, 4 approved, 1 escalated, 1 denied]
}
INTEGRATE
Try every API
Standard REST. JSON in, JSON out. Works with any language. The buttons below send real requests to the OMEGA Decision API using your API key.
Set expectation
Run a decision
Extract assumptions
Record a trace
Agent check
Verify a chain
Compliance report
// Set an expectation baseline before a decision
const exp = await fetch('https://api.omegaprotocol.org/v1/expectation/set', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    session_id: 'ses_trial_2026_03',
    decision_context: 'Phase II go/no-go',
    expected_outcomes: [
      { description: 'Trial proceeds to Phase III', metric: 'primary endpoint met' },
      { description: 'Safety profile acceptable', metric: 'no SAEs' }
    ]
  })
});

// Response
{
  "expectation_id": "exp_...",
  "session_id": "ses_trial_2026_03",
  "expectation_hash": "sha256:..."
}
// Run a governed decision analysis
const decision = await fetch('https://api.omegaprotocol.org/v1/decide/analyze', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    decision: 'Raise Series A at £8M valuation or continue bootstrapping',
    scope: 'strategic',
    posture: 'balanced',
    modes: ['decision_record', 'challenge', 'audit_trace'],
    context: { mrr: 30000, growth: '15% monthly', runway: '18 months' }
  })
});

// Response
{
  "state": "CONDITIONAL",
  "summary": {
    "recommendation": "Defer 3-4 months — growth trajectory suggests better terms ahead",
    "conditions": ["Validate growth sustainability", "Get competing term sheets"],
    "confidence": 0.62
  },
  "assumptions": 8,
  "tensions": 2,
  "trace_hash": "sha256:3e8f1a..."
}
// Extract assumptions from a business plan
const assumptions = await fetch('https://api.omegaprotocol.org/v1/assumptions/extract', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    text: 'We project 15% monthly growth sustained over 12 months, with customer acquisition cost remaining below £40 and no major competitor entering the market.',
    session_id: 'ses_plan_2026'
  })
});

// Response
{
  "assumptions": [
    { "id": "A1", "text": "15% monthly growth sustained for 12 months", "status": "UNTESTED" },
    { "id": "A2", "text": "CAC remains below £40", "status": "UNTESTED" },
    { "id": "A3", "text": "No major competitor enters market", "status": "UNTESTED" }
  ]
}
// Record a hash-chained trace
const trace = await fetch('https://api.omegaprotocol.org/v1/trace/record', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    action: 'Approved compound for Phase II trial',
    context: { domain: 'clinical', confidence: 0.82 },
    session_id: 'ses_trial_2026_03'
  })
});

// Response
{
  "id": "tr_a1b2c3",
  "hash": "sha256:9f4e2d...",
  "previous_hash": "sha256:7c1a8b...",
  "chain_position": 4,
  "integrity": "valid"
}
// Pre-execution check for an agent action
const check = await fetch('https://api.omegaprotocol.org/v1/agent/check', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    action: 'read config from environment',
    session_id: 'ses_agent_2026',
    context: 'Startup validation'
  })
});

// Response
{
  "check_id": "chk_...",
  "session_id": "ses_agent_2026",
  "action": "read config from environment",
  "decision": "APPROVED",
  "reason": "Action is within defined boundaries",
  "flags": [],
  "timestamp": "2026-03-12T..."
}
// Verify an entire trace chain
const verify = await fetch('https://api.omegaprotocol.org/v1/trace/verify', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer om_live_...',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    session_id: 'ses_trial_2026_03'
  })
});

// Response
{
  "session_id": "ses_trial_2026_03",
  "chain_length": 12,
  "integrity": "valid",
  "first_record": "2026-03-01T09:14:22Z",
  "last_record": "2026-03-12T16:45:01Z",
  "tampered": false
}
// Get a framework-specific compliance report for a session
const report = await fetch('https://api.omegaprotocol.org/v1/trace/compliance-report/ses_trial_2026_03?framework=eu_ai_act', {
  headers: {
    'Authorization': 'Bearer om_live_...'
  }
});

// Response
{
  "report_id": "cr_...",
  "session_id": "ses_trial_2026_03",
  "framework": "eu_ai_act",
  "summary": { "total_traces": 12, "chain_integrity": "VALID" },
  "framework_mapping": { "framework_name": "EU AI Act", "articles": [...] }
}
REFERENCE
Complete API Reference
This section documents all seven APIs, authentication, and common error codes so you can integrate the OMEGA Decision API from any stack.
Authentication

All requests to the OMEGA Decision API must include a bearer token:

Authorization: Bearer YOUR_API_KEY
  • Get an API key by entering your invite code in the hero at the top of this page.
  • Keys are prefixed with om_live_.

Plan limits:

  • Free: 1,000 traces / 100 assumptions / 10 decisions per month.
  • Pro: 50,000 traces / unlimited assumptions / 100 decisions per month.
  • Team: unlimited traces, assumptions, and decisions.
Record a trace
POST /v1/trace/record

Record a hash-chained trace entry for any important action or decision. Each call appends a new link to the session's trace chain.

Request headers

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Request body

Field Type Required Description
action string YES The action, event, or decision being traced.
session_id string NO Groups traces into a chain. Auto-generated if omitted.
context object NO Structured context about the action (inputs, environment, user, etc.).
metadata object NO Additional tags and metadata for downstream analytics.
expectation_id string NO Optional link to a prior expectation baseline.
confirmation_id string NO Optional link to a confirmation record.
governance_stage string NO observe | expectation | surprise | derive | assume | decide | confirm | act | no_act | invalidation

Example request body

{
  "session_id": "ses_db_migration_2026_03",
  "action": "DB_MIGRATION_DECISION_RECORDED",
  "context": {
    "decision": "Migrate primary PostgreSQL to CockroachDB overnight",
    "risk_level": "high",
    "team": "platform",
    "window": "02:00–06:00 UTC Saturday"
  },
  "metadata": {
    "lead_engineer": "sara_k",
    "approved_by": "cto"
  },
  "governance_stage": "decide"
}

Example response (201)

{
  "id": "tr_m9x2c4a1b8f3",
  "session_id": "ses_db_migration_2026_03",
  "hash": "sha256:a7f3d92e1b4c8f0e5a2d6b9c3f1e8a4d7b0c5e2f9a6d3b8c1e4f7a0d3b6c9e2f",
  "previous_hash": "sha256:0000000000000000000000000000000000000000000000000000000000000000",
  "chain_position": 1,
  "integrity": "valid",
  "created_at": "2026-03-15T23:14:07.218Z"
}
Extract assumptions
POST /v1/assumptions/extract

Turn free-form text into a structured list of assumptions that can be tracked, tested, and invalidated over time.

Request headers

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Request body

Field Type Required Description
text string YES Source text to scan for implicit or explicit assumptions.
session_id string NO Session to attach assumptions to (for later decisions and traces).

Example request body

{
  "text": "We will migrate the primary database overnight this Saturday. The migration window is 4 hours. We expect zero data loss. Rollback is possible within the first 2 hours if we detect issues. The new system can handle our current query load from day one.",
  "session_id": "ses_db_migration_2026_03"
}

Example response (200)

{
  "session_id": "ses_db_migration_2026_03",
  "assumptions": [
    {
      "id": "A_4f2a9c1e8b3d",
      "label": "A1",
      "text": "Migration will complete within the 4-hour window",
      "status": "UNTESTED"
    },
    {
      "id": "A_7c3b1f9e2a4d",
      "label": "A2",
      "text": "Zero data loss is achievable with the chosen migration approach",
      "status": "UNTESTED"
    },
    {
      "id": "A_1d8e4b2f6c9a",
      "label": "A3",
      "text": "Rollback is possible within the first 2 hours",
      "status": "UNTESTED"
    },
    {
      "id": "A_5a9f3c7d2e1b",
      "label": "A4",
      "text": "New system handles current query load from day one",
      "status": "UNTESTED"
    }
  ],
  "count": 4,
  "extraction_method": "gemini_v1",
  "model_integration": "gemini"
}
Analyze a decision
POST /v1/decide/analyze

Run a structured, multi-lens analysis of an important decision. DECIDE can surface risks, tensions, and recommended actions.

Request headers

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Request body

Field Type Required Description
decision string YES The decision question you want analyzed.
session_id string NO Session to attach this analysis to.
preset string NO Domain preset: clinical, legal, procurement, product, finance, scientific, board.
scope string NO operational | strategic | tactical (defaults are applied if omitted).
posture string NO balanced | conservative | aggressive (affects risk sensitivity).

Example request body

{
  "decision": "Should we migrate our primary PostgreSQL database to CockroachDB this Saturday during the 02:00–06:00 UTC maintenance window?",
  "session_id": "ses_db_migration_2026_03",
  "preset": "architecture",
  "scope": "operational",
  "posture": "conservative"
}

Example response (201)

{
  "id": "dec_9b3f1e7c4a2d",
  "session_id": "ses_db_migration_2026_03",
  "job_id": "job_8f2c1a9e4b3d",
  "decision": "Should we migrate our primary PostgreSQL database to CockroachDB this Saturday during the 02:00–06:00 UTC maintenance window?",
  "scope": "operational",
  "posture": "conservative",
  "state": "PROPOSED",
  "trace_hash": "sha256:c4f8a2d1e9b3f7c0a5d2e8b4f1c7a3d9e6b0f2c8a4d7e1b5f3c9a6d0e4b8f2c5",
  "status": "pending",
  "analysis_status": "running",
  "poll_url": "/v1/decide/job/job_8f2c1a9e4b3d",
  "created_at": "2026-03-15T23:14:22.891Z"
}
Set an expectation baseline
POST /v1/expectation/set

Record what you expect before acting. Expectations are the baseline that surprise and calibration are measured against.

Example request body

{
  "session_id": "ses_db_migration_2026_03",
  "decision_context": "PostgreSQL to CockroachDB migration, Saturday maintenance window",
  "expected_outcomes": [
    {
      "dimension": "migration_duration_hours",
      "expected_value": "4",
      "confidence": 0.65,
      "basis": "staging_environment_test",
      "testable_by": "migration_completion_timestamp",
      "horizon": "2026-03-21T06:00:00Z"
    },
    {
      "dimension": "data_loss_records",
      "expected_value": "0",
      "confidence": 0.88,
      "basis": "checksums_and_row_counts",
      "testable_by": "post_migration_audit_query",
      "horizon": "2026-03-21T08:00:00Z"
    },
    {
      "dimension": "query_performance_p99_ms",
      "expected_value": "under 200",
      "confidence": 0.55,
      "basis": "vendor_benchmarks",
      "testable_by": "production_query_monitoring",
      "horizon": "2026-03-22T00:00:00Z"
    }
  ],
  "baseline_state": {
    "description": "Pre-migration state: PostgreSQL p99 query time 145ms, 2.3M rows in primary table, last full backup 18:00 UTC Friday",
    "key_variables": {
      "current_p99_ms": 145,
      "primary_table_rows": 2300000,
      "last_backup": "2026-03-20T18:00:00Z"
    }
  }
}
Record a confirmation gate
POST /v1/confirmation/record

Separate DECIDE from ACT. Confirmation records capture whether intent was confirmed, failed, or expired — including non-actions.

Example request body

{
  "session_id": "ses_db_migration_2026_03",
  "expectation_id": "exp_20260320_db_migration",
  "intent": {
    "description": "Begin live database migration — cut over production traffic to CockroachDB",
    "action_type": "infrastructure_change",
    "provisional_at": "2026-03-21T01:30:00Z"
  },
  "confirmation_window_seconds": 1800,
  "confirmation_checks": [
    {
      "check": "staging_migration_passed",
      "result": "pass",
      "observed_value": "completed in 3h 42m, zero errors",
      "expected_value": "completed within window"
    },
    {
      "check": "backup_verified",
      "result": "pass",
      "observed_value": "full backup confirmed at 18:00 UTC",
      "expected_value": "backup within 12 hours"
    },
    {
      "check": "rollback_tested",
      "result": "pass",
      "observed_value": "rollback tested in staging, 8 minutes",
      "expected_value": "rollback under 30 minutes"
    },
    {
      "check": "on_call_engineer_available",
      "result": "pass",
      "observed_value": "sara_k and backup on-call confirmed",
      "expected_value": "two engineers available"
    }
  ],
  "confirmation_status": "confirmed",
  "confirmation_score": 0.95,
  "acted": true,
  "reason": "All four confirmation checks passed. Proceeding with migration at 02:00 UTC."
}
Score surprise against an expectation
POST /v1/surprise/score

When outcomes resolve, compute a surprise score against the original expectation baseline. Use this for calibration and model governance.

Example request body

{
  "expectation_id": "exp_20260320_db_migration",
  "actual_outcomes": [
    {
      "dimension": "migration_duration_hours",
      "actual_value": "9.2",
      "observed_at": "2026-03-21T11:12:00Z"
    },
    {
      "dimension": "data_loss_records",
      "actual_value": "0",
      "observed_at": "2026-03-21T11:30:00Z"
    },
    {
      "dimension": "query_performance_p99_ms",
      "actual_value": "380",
      "observed_at": "2026-03-21T14:00:00Z"
    }
  ]
}

Example response (200)

{
  "expectation_id": "exp_20260320_db_migration",
  "surprise_score": 0.67,
  "surprise_by_dimension": [
    {
      "dimension": "migration_duration_hours",
      "expected": "4",
      "actual": "9.2",
      "divergence": "high",
      "direction": "above"
    },
    {
      "dimension": "data_loss_records",
      "expected": "0",
      "actual": "0",
      "divergence": "low",
      "direction": "aligned"
    },
    {
      "dimension": "query_performance_p99_ms",
      "expected": "under 200",
      "actual": "380",
      "divergence": "high",
      "direction": "above"
    }
  ],
  "calibration_note": "Significant divergence between expectation and outcome — migration took 2.3× longer than expected and query performance fell short. Review assumptions A1 and A4.",
  "assumption_failures": [
    "Expectation on migration_duration_hours was invalidated by the outcome.",
    "Expectation on query_performance_p99_ms was invalidated by the outcome."
  ]
}
Poll a decision job
GET /v1/decide/job/:jobId

DECIDE analyses run asynchronously. Poll this endpoint until the job status is complete.

Example

GET /v1/decide/job/job_206fbfdfb393
Authorization: Bearer YOUR_API_KEY

Example response when complete (200)

{
  "job_id": "job_8f2c1a9e4b3d",
  "status": "complete",
  "result": {
    "summary": "Migration is viable but high-risk. Staging tests are encouraging but four assumptions remain untested in production conditions. Recommend confirming rollback procedure before proceeding.",
    "outputs": [],
    "assumptions_surfaced": 4,
    "trace_hash": "sha256:c4f8a2d1...",
    "expectation_baseline": {
      "what_is_assumed": [
        "Migration completes within 4-hour window",
        "Zero data loss achievable",
        "Rollback possible within 2 hours",
        "New system handles current query load"
      ],
      "what_would_surprise": [
        "Migration exceeds 6 hours",
        "Any data integrity failures detected",
        "Rollback window closes before issues detected"
      ],
      "confirmation_checks": [
        "Verify staging migration completed cleanly",
        "Confirm backup within last 12 hours",
        "Test rollback procedure before go-live"
      ]
    },
    "primitive_coverage": {
      "governance": true,
      "reasoning": true,
      "traceability": true,
      "expectation": true,
      "confirmation": false
    }
  }
}
Verify a trace chain
GET /v1/trace/chain/:session_id

Return the full hash chain for a session to verify integrity end-to-end. Use this to audit every decision and action in a session.

Example response (200)

{
  "session_id": "ses_db_migration_2026_03",
  "chain": [
    {
      "id": "tr_m9x2c4a1b8f3",
      "action": "DB_MIGRATION_DECISION_RECORDED",
      "hash": "sha256:a7f3d92e...",
      "previous_hash": "sha256:00000000...",
      "chain_position": 1,
      "governance_stage": "decide",
      "integrity": "valid"
    },
    {
      "id": "tr_k4p7e1n9q2r5",
      "action": "expectation_set",
      "hash": "sha256:b8e4c1f9...",
      "previous_hash": "sha256:a7f3d92e...",
      "chain_position": 2,
      "governance_stage": "expectation",
      "integrity": "valid"
    },
    {
      "id": "tr_r3s6t9u2v5w8",
      "action": "confirmation_recorded",
      "hash": "sha256:c9f5d2a8...",
      "previous_hash": "sha256:b8e4c1f9...",
      "chain_position": 3,
      "governance_stage": "act",
      "integrity": "valid"
    }
  ],
  "length": 3,
  "integrity": "valid"
}
Compliance report
GET /v1/trace/compliance-report/:sessionId?framework=eu_ai_act

Generate a structured compliance report from trace, assumption, and decision data for a session. Supported framework query values: eu_ai_act, iso_42001, nist_ai_rmf, general (default), fda_21cfr11 (placeholder).

Trace stream (SSE)
GET /v1/trace/stream/:sessionId

Server-Sent Events endpoint for real-time trace updates. Sends existing traces first, then polls and pushes updates every 2 seconds. Use ?api_key= for browser EventSource when needed.

Dashboard
GET /v1/dashboard/:sessionId

HTML dashboard for a session: traces (newest first), assumption registry, decisions, chain integrity, and a "Generate Compliance Report" button. Auto-updates via SSE. Use ?api_key= for browser access.

Agent pre-execution check
POST /v1/agent/check

Submit a proposed agent action. The API checks it against boundaries and returns APPROVED, DENIED, or ESCALATE. Optional Gemini risk assessment. Body: action (required), session_id (required), agent_id, context, boundaries.

Error codes

Common HTTP statuses and error codes returned by the OMEGA Decision API.

Status Code Meaning
400 bad_request Missing or invalid fields in the request body.
401 unauthorized Missing, invalid, or expired API key.
404 not_found The requested resource does not exist.
409 code_already_used This invite code has already been redeemed.
429 usage_limit_exceeded Monthly plan limit reached — upgrade to continue.
500 internal_error Unhandled server error — retry or contact support.
USE CASES
Built for high-stakes domains
Any system where decisions are irreversible, accountability is distributed, or assumptions drift over time.

Scientific research

Trace experimental decisions, surface hidden assumptions, and keep a reproducible reasoning trail alongside your lab notebook and data.

AI agents & Agent Guard

Pre-execution checks for autonomous agents. Boundary enforcement, risk assessment, and escalation gates. Hash-chained traces and verifiable audit chains for every action an agent proposes.

Clinical decisions

Decision records and assumption registries for trial design, diagnostics, and treatment choices that need to withstand regulatory review.

Autonomous systems

For robots, vehicles, drones, and spacecraft. Capture every control decision as a trace with forensic replay and safety checks.

Regulated industries

Generate evidence for EU AI Act, FDA, ISO 42001, and NIST frameworks directly from trace and assumption data.

Compliance & Audit

Generate framework-specific compliance reports from your trace data. EU AI Act, ISO 42001, NIST AI RMF.

Founders and boards

Structured strategic decisions with tracked assumptions, exportable governance records, and replayable board histories.

Defence and national security

Zero-dependency, air-gapped compatible decision tracing for systems that cannot rely on external services.

Financial services

Inspectable underwriting, trading, and credit decisions. Assumption registries for model risk management.

COMPOUND INTELLIGENCE
Every decision makes the next one better.

Most tools reset every time you use them. OMEGA remembers.

Day 1
Your first traced decision

Record a decision. Extract assumptions. Set an expectation. You have a baseline.

Month 1
Patterns emerge

50 decisions traced. Your assumption registry shows which assumptions keep failing. Surprise scores reveal where your predictions are weakest. You start making better calls.

Month 6
Institutional memory

500 decisions. Full calibration history. You can see which types of decisions your team consistently gets wrong, which assumptions are always untested, and which domains carry the most surprise. New team members inherit the organisation's decision intelligence from day one.

Year 1
Unreplicable advantage

Thousands of traced decisions with hash-chained integrity. Compliance reports generated instantly. No competitor can replicate your decision history. The data compounds — and it's yours.

The longer you use it, the more it knows. The more it knows, the better your decisions. That's compound intelligence.

SECURITY & TRUST
Built for data you can't afford to lose.
Encrypted in transit

All API communication over HTTPS/TLS. No plaintext. No exceptions.

Scoped API keys

Every key is unique, plan-scoped, and revocable. Keys are never logged or exposed in responses.

Tamper-evident traces

Every record is SHA-256 hash-chained. If any record is altered, the chain breaks and verification fails. Immutable by design.

Your data stays yours

Session data is isolated per API key. No cross-tenant access. We do not use your data to train models or share it with third parties.

UK & EU hosted

Infrastructure hosted in European data centres. Aligned with GDPR and UK Data Protection Act 2018. Enterprise plan includes dedicated instance and data residency options.

Open protocol

The OMEGA Protocol is MIT licensed and independently verifiable. You can audit the trace format, hash algorithm, and data model yourself. No vendor lock-in.

SOLUTIONS
One engine. Many domains.
Choose a preset to configure scope, posture, analytical modes, and output format in one go. The preset parameter wires DECIDE for your domain without changing your code.

Board Brief

Summarise strategic decisions into a board-ready memo with assumptions and tensions explicitly called out.

Due Diligence

Stress-test an investment or acquisition thesis with multi-lens analysis and an auditable assumption registry.

Grant Evaluation

Structure panel decisions, surface hidden premises, and export a consistent record for every grant round.

Architecture Review

Run technical design reviews with explicit risks, mitigations, and traceable decisions across services and systems.

Incident Forensics

Reconstruct what happened after an outage or failure with replayable traces and a structured incident report.

Hypothesis Engine

Generate, track, and test research hypotheses with a living assumption registry across trials and experiments.

Procurement Review

Govern vendor and tooling choices with explicit criteria, challenges, and a clean audit trail for approvals.

Custom preset

Define your own preset for any domain, then reuse it across teams and environments.

PRICING
Start free. Scale when ready.
Every plan includes all seven APIs: Trace, Assumptions, Decide, Expectation, Confirmation, Surprise, and Agent. One API key gives you access to the full governance spine.
Deep Science Ventures · Living Proof
VSD venture scientists and Living Proof members get Pro free. Trace your experiments, track your assumptions, govern your venture decisions from day one.
Free
£0
Evaluate and prototype
1,000 traces/month
100 assumption extractions
10 decision analyses
10 expectation baselines
10 confirmation records
Community support
Pro
£99 /month
For teams shipping governed AI
50,000 traces/month
Unlimited assumptions
100 decision analyses
500 expectation baselines
500 confirmation records
Export + compliance reports
Priority support
Team
£499 /month
For organisations and portfolios
High-volume usage
Unlimited expectations + confirmations
Team dashboard
SSO + audit logs
Compliance reports
Dedicated support
Enterprise
Custom
On-premise, SLAs, custom integration
Self-hosted option
Custom SLAs
Dedicated instance
Integration support
Contact us
8
Protocols
297
Tests passing
5
Live demos
0
Dependencies
<10ms
Core API latency
74%
Deploying AI agents within 2 years
21%
Have governance in place
Reference implementation metrics from the current protocol stack. Market data: Deloitte State of AI in the Enterprise, 2026.

Open protocol. Managed infrastructure.

OMEGA Protocol is MIT licensed and independently verifiable. The APIs are the hosted, managed layer on top.

View protocols