Your AI has no governance layer.
Fix that in one API call.
DreamerOS is not another AI wrapper. It is a governed execution pipeline. Every response passes through intent expansion, integrity checking, drift detection, and follow-up forecasting before it reaches your application. Ship AI features that are structured, verified, and auditable from day one.
from dreameros import DreamerOS
client = DreamerOS(api_key="dos_...")
# One call. Governed output.
response = client.prompt(
input="Summarize the key risks in this contract",
mode="structured",
governance=True
)
# Every response includes:
# - Governed output (EDE structured)
# - DAIM integrity score
# - Digital Buoyancy breakdown
# - Session ID for memory
# - Routing metadata
print(response.output)
print(response.integrity_score) # 0.97
print(response.governance.drift) # NoneThis is the developer surface. If you do not write code, you do not need anything on this page - the same governance runs invisibly when you use the Pro web app or the GPT. If you do write code, this is how you ship the same pipeline inside your own product: one HTTP call, structured output, an integrity score on every response, a session id for memory, and an audit trail your compliance team can actually export. The endpoints below cover the full pipeline. The quickstart below them runs in three lines.
What happens inside every API call
Six stages. All invisible to the end user. All auditable by your team.
Every stage is logged. Every stage is queryable. Your compliance team will love you.
Six endpoints. Every response governed.
REST API. JSON in, JSON out. Bearer auth. Rate limits scale with your tier.
The core endpoint. Send natural language, get structured intent-verified output. Every response passes through the full EDE + DAIM + Routing pipeline.
EDE structured DAIM verified Auto-routed{
"output": { "quick": "...", "deep": "...", "forecast": "..." },
"integrity": { "score": 0.97, "flags": [], "drift": null },
"governance": { "auth": 100, "authz": 98, "acct": 100 },
"routing": { "engine": "execution", "mode": "strategic" },
"session_id": "ses_a1b2c3..."
}Submit any AI output for integrity verification. Returns hallucination flags, assumption detection, unsupported claims, and a confidence score. Works on output from any model, not just DreamerOS.
Hallucination scan Assumption flags{
"score": 0.94,
"flags": [{ "type": "assumption", "text": "Assumes US jurisdiction" }],
"silent_drops": [],
"drift_detected": false
}Let DreamerOS pick the optimal engine and thinking mode for a task. Returns routing metadata without executing the prompt. Useful for pre-flight checks and custom pipelines.
auto mode detection 6 core domainsRetrieve session state, conversation history, and accumulated context. Sessions persist across API calls so every subsequent request builds on prior interactions.
Persistent memory Cross-call contextRegister webhook URLs for real-time event callbacks. Get notified when tasks hit governance checkpoints, chains complete, integrity scores drop below threshold, or human approval is needed.
Review gates Automation triggersDigital Buoyancy compliance score for your account. Authentication, authorization, and accounting metrics aggregated across all API calls. Exportable for compliance reporting.
Compliance ready Audit trailTen endpoints. Governance and trust primitives for AI products.
Built on the core pipeline. Verify any AI output, underwrite a claim, issue bonds against governance scores, prove provenance, and export compliance audit trails. Five public-callable consumer endpoints, five B2B and enterprise endpoints. All authenticated with the same Bearer token. All billed transparently through governance_events.
Submit any text claim for governance verification. Runs DAIM + IFP + EDE + CTCI and returns a signed verdict you can share with your own customers. Three depth tiers: light (free, any tier), deep (Pro+, $0.05), exhaustive (Elite, $0.25). Response detail scales with depth.
light: free deep: $0.05 exhaustive: $0.25{
"verdict": "verified",
"score": 0.94,
"depth": "deep",
"flags": [],
"signature": "ed25519:...",
"cost_usd": 0.05
}Export your cognitive fingerprint as a signed portable passport. Five behavioral axes plus engine trust scores, signed with CTCI ed25519. Any downstream AI can verify it and import it via MCP, so your tone, preferences, and trust history follow you to every tool you use.
Cross-tool identity CTCI-signed MCP-importablePublic, no-auth feed of the top 20 users by high and critical governance catches in the past 7 days. User IDs are SHA-256 pseudonyms - no PII. The point: prove DAIM and IFP are catching real drift in real time, not just running benchmarks.
Public PseudonymousUnified event feed across Sentry, Railway, Vercel, Supabase, and GitHub for your account. One endpoint, one shape, one auth. Used by Command Center to surface "what just broke" without you opening five dashboards. Per-source failures degrade to a warning - never blank the whole feed.
Elite 5 sources unifiedSend an ambiguous prompt, get back the EDE-restructured intent the gateway would have actually run. Public-callable for any tier; Pro+ unlocks the full intent layer (emotion signal, depth hint, ambiguity score). Use it to pre-flight a prompt before paying for execution, or to debug why an AI misunderstood you.
Public Pro+ fullSubmit an AI-generated decision or recommendation; get back an underwritten verdict you can stand behind. Combines IFP score, depth audit, and a signed certificate of underwriting. Pro+ at $0.10 per call. Use it when "I asked the AI" is not a defensible answer.
Pro+ $0.10 / call CTCI-signedIssue and list bonds backed by governance scores. GET /listings returns public bond inventory. POST /issue mints a new bond against an underwritten claim. Think Lloyds of London for AI outputs - bonded performance instead of unbacked promises. Pro+ to issue, public to read.
{
"bond_id": "bond_a1b2c3...",
"backing_score": 0.96,
"underwriter_signature": "ed25519:...",
"face_value_usd": 10000
}Anchor an AI output to a verifiable provenance chain. POST /claim registers a new claim with engine, model, governance metadata, and a content hash. GET /{claim_id} resolves the chain. C2PA-aligned. Use it to prove which model said what, when, and under which governance gate.
IFP Adherence Audit Export. Returns the full append-only governance_events trail for your account in a compliance-ready bundle (JSONL + manifest + signature). One endpoint, zero scraping, ready for SOC, ISO, and internal review. Elite only.
Elite Append-only Compliance-readyTier 2 endpoints are usage-billed through governance_events.cost_usd - no separate metered-billing integration. Costs land on your Stripe invoice at month-end, itemized by endpoint. Underwriter is $0.10 per call, Bonds is per-bond face-value percentage, Provenance is included with Pro+, Audit Export is included with Elite. No surprise charges - every call is logged before it bills.
Governed output inside any AI tool
DreamerOS is available as an MCP (Model Context Protocol) connector for any compatible AI client. Plug DreamerOS governance directly into Claude, ChatGPT, Gemini, Grok, Perplexity, or any MCP-compatible tool.
Light connector (free)
Behavioral governance. The 27-principle contract shapes output quality without computational transformation. Available for all supported engines.
Pro connector ($29.99/month)
Full Trinity pipeline. EDE restructuring, DAIM validation, smart routing, token efficiency tracking, and connected memory. The complete governance layer running inside your preferred AI tool.
Integration
The connector endpoint is live at the DreamerOS gateway. Authentication is via API key. Tier-based routing is handled server-side - the connector surface is identical for Light and Pro, the pipeline depth changes based on your subscription.
Three lines to intent-verified AI
Python SDK, Node SDK, or raw cURL. Pick your weapon.
Python
Async-ready. Type-hinted. Works with FastAPI, Django, Flask.
Node.js
ESM and CommonJS. Works with Next.js, Express, serverless.
cURL / REST
Standard REST. JSON in, JSON out. Bearer auth. Works with anything.
from dreameros import DreamerOS
client = DreamerOS(api_key="dos_your_key")
response = client.prompt(
input="Should I open a second location?",
governance=True
)
print(response.output.quick)
# "Before expanding, audit unit economics on location one.
# Most second-location failures are location-one problems
# that got exported."
print(response.integrity_score)
# 0.96 - no hallucinations, no assumptions flaggedGovernance that ships with your code
Drift detection in your pipeline. Canon verification on every deploy. Integrity scoring in your tests.
Drift Sentinel
Automated behavioral tests that run on every deploy. Define expected behaviors in YAML. DreamerOS runs test prompts against the live system and verifies outputs match your baselines. Drop below 90% pass rate and the pipeline halts.
# .github/workflows/drift.yml
- name: Run Drift Sentinel
run: dreameros drift-test --baseline v5.yamlCanon Verification
Every governance file is SHA-256 hashed in a registry. The gateway verifies hashes on startup. If any file has been tampered with, the system refuses to start. Add verify_canon.py to your CI to catch drift before it ships.
# Block deploys on canon mismatch
- name: Verify Canon Integrity
run: python verify_canon.pyFail-Closed Gateway
If DAIM is unreachable, the API returns 503, not a degraded response. If canon verification fails, the gateway refuses to start. If direction sets fail to load, no requests are served. Governance is never optional.
# No silent degradation. Ever.
"error": "governance_unavailable"
"status": 503Every request authenticated, authorized, accounted for
API keys are scoped to your organization and tied to tier permissions. Rate limits scale with your plan. Every request is logged under the Digital Buoyancy accounting layer.
Keys are generated from your DreamerOS dashboard. Each key is bound to a single organization with configurable permission scopes. Keys can be rotated, revoked, or scoped to specific endpoints at any time.
Rate limits are applied per key and vary by tier - from development sandbox to production throughput. All requests are logged for full auditability.
# Bearer token authentication
curl -X POST https://api.dreameros.app/v1/prompt \
-H "Authorization: Bearer dos_your_key" \
-H "Content-Type: application/json" \
-d '{
"input": "Analyze this quarterly report",
"governance": true
}'
# Response headers include:
# X-RateLimit-Remaining: 94
# X-Governance-Score: 98
# X-DAIM-Status: passed
# X-Session-ID: ses_a1b2c3Builders who take AI seriously
Whether you are shipping your first AI feature or governing AI at organizational scale.
Solo devs and indie hackers
Ship AI features with built-in governance from day one. Hallucination detection, integrity checking, and audit trails come free with every API call. Focus on your product, not AI reliability engineering.
Product teams and SaaS platforms
Embed DreamerOS into your existing product. Give your users AI-powered features that are reliable and accountable. Every response includes governance metadata so you can build trust into your UX.
CS students and AI researchers
DreamerOS exposes its governance internals. Every response includes routing metadata, integrity scores, and engine selection rationale. Use it to study how intent-verified AI systems work, build research projects around drift detection, or benchmark integrity scoring against your own models.
Enterprise and Elite teams
Multi-engine orchestration, organization-wide session memory, role-based API keys, full audit logging, custom domain modules, and webhook-driven automation. The governance infrastructure you would build yourself, already built and battle-tested.
Ready to build?
Request API access and start integrating Digital Buoyancy governance into your applications.
Or see all tiers side by side for the full feature breakdown.