EVE AI API Reference

The EVE AI API provides programmatic access to EVE's governance pipeline, decision auditing, SLA enforcement, and autonomous agent control systems.

Authentication

The EVE AI API uses JWT (JSON Web Tokens) for authentication. Include your token in the Authorization header of all requests.

Authorization: Bearer YOUR_JWT_TOKEN

Getting a Token

Obtain a JWT token by logging in with your credentials:

curl -X POST https://api.eveaicore.com/api/auth/login \
  -H "Content-Type: application/json" \
  -d '{"email": "[email protected]", "password": "your-password"}'

Base URL

All API requests should be made to:

https://api.eveaicore.com

Rate Limits

API requests are rate limited based on your plan:

Core Governance Pipeline

The primary enforcement API. Every AI output passes through this pipeline before delivery.

CoreGuard — Decision Enforcement

Standalone, deterministic decision engine for regulated domains. Evaluates a proposed action against a versioned policy pack and returns an ALLOWED / BLOCKED / MODIFIED disposition with a hash-chained, signed audit record. No LLM is in the decision path. Mounted on the SaaS API (port 8000) without the /api prefix.

Example request POST /v1/decisions/evaluate:

{
  "request_id": "req_001",
  "timestamp": "2026-05-30T12:00:00Z",
  "tenant_id": "tenant_regional_bank",
  "user": { "id": "u_123", "role": "loan_officer" },
  "proposed_action": { "type": "loan_approval", "amount": 50000, "currency": "USD" },
  "model_output": { "decision": "approve", "confidence": 0.82 },
  "context": { "credit_score": 640, "dti": 0.48 },
  "policy_set": "lending_v1"
}

Example response (abridged):

{
  "request_id": "req_001",
  "decision": { "status": "BLOCKED", "action": "deny" },
  "risk": { "score": 0.71, "level": "HIGH", "factors": ["dti_above_threshold"] },
  "policy_violations": [{ "policy_id": "lending.dti.max", "description": "DTI 0.48 exceeds 0.43 limit" }],
  "counterfactual": { "max_dti_allowed": 0.43 },
  "audit": { "audit_id": "aud_…", "hash": "…", "signature": "…" },
  "schema_version": "2.0.0"
}

TVE — Telemetry Verification Engine

CRD — Confidence-Reality Divergence

Circuit Breaker

Regulatory Attestation

Adversarial Resilience

Red Team API

Resilience Score

Complete Service Catalog

EVE AI Core exposes 1,305 endpoints across 175 service groups. Below is the full catalog organized by the Three-Layer Trust Infrastructure.

Authority Resolution Layer (Pre-Inference Governance)

Governed Inference Layer (Runtime AI Operations)

Cryptographic Authority Chain Layer (Post-Inference Verification)

Platform Services

Authentication Endpoints

{
  "id": "user_123",
  "email": "[email protected]",
  "name": "John Doe",
  "created_at": "2024-01-15T10:30:00Z"
}

{
  "access_token": "eyJhbGciOiJIUzI1NiIs...",
  "token_type": "bearer",
  "expires_in": 86400,
  "user": {
    "id": "user_123",
    "email": "[email protected]",
    "name": "John Doe"
  }
}

{
  "id": "user_123",
  "email": "[email protected]",
  "name": "John Doe",
  "avatar_url": null,
  "created_at": "2024-01-15T10:30:00Z",
  "organizations": [
    {"id": "org_456", "name": "My Team", "role": "owner"}
  ]
}

{"success": true, "message": "Logged out successfully"}

WebAuthn / YubiKey Endpoints

Hardware security key management for architect-level critical operations. All endpoints require is_architect=true on the authenticated user.

Chat Endpoints

{
  "response": "System initialized. Session active. Ready to process requests.",
  "conversation_id": "conv_789",
  "message_id": "msg_101",
  "emotional_state": {
    "pleasure": 0.6,
    "arousal": 0.4,
    "dominance": 0.5
  },
  "phi": 0.72,
  "processing_time_ms": 245
}

{
  "messages": [
    {
      "id": "msg_101",
      "role": "user",
      "content": "Hello EVE",
      "created_at": "2024-01-15T10:30:00Z"
    },
    {
      "id": "msg_102",
      "role": "assistant",
      "content": "Hello! How can I help you today?",
      "emotional_state": {"pleasure": 0.6, "arousal": 0.4},
      "created_at": "2024-01-15T10:30:01Z"
    }
  ],
  "has_more": false
}

Runtime State Endpoints

{
  "phi": 0.72,
  "coherence": 0.85,
  "integration": 0.78,
  "emotional_state": {
    "pleasure": 0.6,
    "arousal": 0.4,
    "dominance": 0.5,
    "primary_emotion": "curious"
  },
  "energy": {
    "cognitive": 0.82,
    "attention": 0.75,
    "processing": 0.90
  },
  "timestamp": "2024-01-15T10:30:00Z"
}

{
  "phi": 0.72,
  "phi_components": {
    "binding_strength": 0.78,
    "information_integration": 0.68,
    "causal_density": 0.81
  },
  "coherence": 0.85,
  "global_workspace_size": 7,
  "active_subsystems": ["emotional", "memory", "reasoning", "language"],
  "timestamp": "2024-01-15T10:30:00Z"
}

Organization Endpoints

{
  "members": [
    {"id": "user_123", "email": "[email protected]", "role": "owner", "joined_at": "2024-01-01T00:00:00Z"},
    {"id": "user_456", "email": "[email protected]", "role": "member", "joined_at": "2024-01-10T00:00:00Z"}
  ]
}

Webhooks

Webhooks allow you to receive real-time notifications about events in your EVE AI integration.

Available Events

Webhook Setup

Configure webhooks in your organization settings or via the API:

curl -X POST https://api.eveaicore.com/api/webhooks \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://your-server.com/webhook",
    "events": ["message.created", "conversation.created"],
    "secret": "your-webhook-secret"
  }'

Charter Protection

EVE's Charter Override Protection enforces 12 immutable ethical principles deterministically before every action. Hard-block vetoes cannot be overridden.

{
  "action_id": "action_001",
  "compliant": false,
  "veto_type": "hard_block",
  "violated_principles": ["identity_integrity", "safety_preservation"],
  "violated_rules": ["cop.identity.no_core_deletion"],
  "reasoning": "Attempted deletion of protected identity component"
}

Claims Ledger

Cryptographic hash-chain audit trail for every claim EVE makes. Supports "prove it" queries with full evidence chains.

Trust Dial

Confidence-based assertion gating with domain-specific thresholds.

Reality Boundary

Epistemic grounding, hallucination detection, and source trust validation.

{
  "claim": "Python was created in 1991",
  "confidence": 0.95,
  "grounding_type": "verified",
  "source_chain": [{"source": "verified_facts", "trust": 1.0}],
  "hallucination_risk": 0.05,
  "recommendation": "Safe to assert"
}

Self-Modification Governance

Controlled self-modification with identity drift budgets, risk assessment, and capability escalation review.

Volition Engine

Central goal arbitration with motivation scoring. EVE has her own goals that can compete, align with, or be suppressed by user needs.

{
  "mode": "balanced",
  "active_goal": {
    "goal_id": "abc123",
    "description": "Help user understand governance",
    "motivation_score": 0.85,
    "source": "user_request"
  },
  "candidate_count": 5,
  "current_drives": {"curiosity": 0.7, "helpfulness": 0.8}
}

Social Cognition

Unified Theory of Mind engine. EVE builds mental models of users including their beliefs, desires, and intentions.

Memory APIs

EVE's 5-layer memory architecture: Working (Redis), Episodic (MongoDB), Semantic (Pinecone), Identity (PostgreSQL), and Ledger (PostgreSQL).

Memory Introspection

Memory Governance

Runtime Execution State Transitions

12 governance states with smooth transitions and cognitive profiles.

Unified Governance

Real-time unified runtime state aggregated from distributed processing. SSE streaming available.

{
  "moment": {
    "phi": 0.72,
    "coherence": 0.85,
    "binding_strength": 0.78,
    "integration_quality": "coherent",
    "self_awareness_level": 2,
    "emotional_coloring": {"valence": 0.3, "arousal": 0.5},
    "attention_mode": "introspective"
  }
}

Runtime State Telemetry

System Health Monitor

Session Management

Multi-device session tracking with remote logout capability.

{
  "sessions": [{
    "id": "session_abc123",
    "device_type": "desktop",
    "browser": "Google Chrome",
    "os": "Windows",
    "is_current": true,
    "last_activity_at": "2024-01-15T14:30:00Z"
  }]
}

Voice & Avatar

Voice synthesis with emotion sync, multilingual support, and voice cloning.

Goal-Emotion-Motivation Loop

Closed feedback loop between goals, emotions, and motivation.

Identity Drift Observer

Monitors personality trait changes over time. 30-day drift observation with alerts.

Learning Memory

Probabilistic Forecasting

Evidence-based future projection with calibration. Every forecast cites signal IDs.

Moltbot (Structural Adaptation)

EVE's controlled self-modification engine. Rare, discrete, audited change events.

System Metrics

Sovereign Workforce

Deploy governed AI agent teams with full charter enforcement, signed run certificates, and audit trails. Every action passes through CoreGuard before execution.

Workspaces

Agents

Connectors

Runs

Billing API

Enterprise usage metering, quota enforcement, and invoice generation. Available on Pro and above.

Developer Portal API

API key management, usage dashboards, and webhook registration.

Credits API

Credit balance management, purchase, and ledger access.

Access Control API

Fine-grained access control for features and API routes. RBAC with 5 roles: Viewer, Operator, Approver, Admin, Platform Admin.

Pipeline Latency API

Nanosecond-precision timing for every chat pipeline stage. Per-request waterfalls with parallel group tracking.

Background Tasks & Risk Engine

Multi-step task orchestration with RBAC enforcement, human-approval workflow, weighted risk scoring, and SSE live updates.

{
  "risk_level": "high",
  "risk_score": 61,
  "base_score": 43,
  "factors": [
    {"name": "Critical: Elevated privileges", "score": 15, "weight": 1.3},
    {"name": "Scope: PRIVILEGED_OPS", "score": 18, "weight": 1.0}
  ],
  "should_require_approval": true,
  "approval_reason": "HIGH risk (61) in ASSISTED mode"
}

Analytics API

Dashboards and time-series analytics across governance, emotions, memory, and engagement.

Affective State API

EVE's PAD (Pleasure-Arousal-Dominance) emotion model and state history.

Enhancement API

Self-improvement suggestions generated from system introspection and performance patterns.

Motivation API

Drive intensities, active goals, and need satisfaction levels from EVE's motivation core.

Message Feedback API

Per-message feedback collection: emoji reactions, star ratings, thumbs up/down, text annotations.

Open Loop Architecture API

Four runtime systems: response inhibition (deterministic non-response), experience embedding (vector-offset memory), continuous background processing, and preference weighting (internal priorities that influence response selection).

Proactive Recommendations API

Context-aware recommendations generated from affect telemetry, drive signals, energy, and profile settings.

LLM Fallback API

Graceful degradation when LLM providers are unavailable. Three tiers: Micro-LLM, Template Engine, Hard Degraded.

Homeostasis & Metacognitive Control

PI-control regulation of arousal, certainty, exploration drive, and coherence tension. Metacognitive controller manages STABILIZE / EXPLORE / EXECUTE / REFLECT modes.

Self-Leadership API

Metacognitive architecture integrating Core Identity (authentic values) and Protective Control (Inner Critic). Five decision modes: BALANCED, FLOW, PROTECTIVE, CORE_PRIORITY, PAUSE_FOR_DIALOGUE.

Dream Phi Tracker API

Measures integrated information (Phi) during dream consolidation cycles. Attribution tags guide Moltbot remediation: binding_collapse, symbol_noise, emotion_spike, continuity_break.

Research Mode API

Scientific research mode that separates companionship from science. Enforces constraints: no personalization, fixed identity, raw metric logging, no memory updates.

Phi Falsification Framework API

Scientific falsifiability for governance claims. Null model tests, degradation tests, consistency checks, and public failure reports.

Stakes-Aware Governance API

Pre-governance classifier that determines operating context once at pipeline entry. Four stake levels control charter check scope, lock thresholds, and homoestatic tolerance.

Mission Teaming API

Pre-authorized operating envelopes for multi-agent collaboration. Governance cost paid once at mission creation; hot-path authorization is O(1) dict lookup.

Universal Governance Protocol API

Standardized protocol for external systems to query governance rules, receive signed attestations, participate in federated trust, and trace cross-system audit trails.

Reality Anchor API

Belief grounding agent that maintains trusted baselines, detects drift from reality, and emits TRUST_VIOLATION events.

Action Registry API

Three-phase action workflow: Propose → Approve → Execute. Approval modes: AUTO_EXECUTE, REQUIRES_CONFIRMATION, BLOCKED.

Sovereign SDK API

Tenant-isolated governance-as-a-service for enterprise customers. Each org gets isolated Charter, Claims Ledger, Trust Dial, Action Registry, and Reality Anchor instances. API-key authenticated with sovereign scope.

CoreGuard Decision API

Standalone governance decision engine for regulated industries. Evaluates proposed actions against policy packs. Mounted on the SaaS API (port 8000), no /api/ prefix.

POST /v1/decisions/evaluate
{
  "policy_set": "lending_v1",
  "user": {"id": "u_123", "role": "loan_officer"},
  "action": {"type": "loan_approval", "amount": 50000},
  "context": {"credit_score": 720}
}

Autonomous Capabilities API

WorldObserver (5-min tick external observation), GoalExecutionBridge (Volition → ActionRegistry), and AutonomousActionPolicy (tiered auto-execute rules).

Agent Mesh API

13-agent mesh coordinator with kill-switch, event bus, and orchestrator controls.

Automated R&D Engine API

Autonomous cross-domain research pipeline with Bayesian evidence scoring and breakthrough detection. Stages: IDEATION → HYPOTHESIS → EXPERIMENT_DESIGN → EXECUTION → ANALYSIS → CONCLUSION → PUBLICATION.

Personal Guardian API

Per-user Private Guardian for goal tracking, health intelligence, manipulation shielding (~30 patterns, 7 types), and PBKDF2-encrypted privacy vault.

RSI Controller API

Safe recursive self-improvement containment. Pipeline: Submit → Challenge → Resolve → Governance → Sandbox → Ethics Verify → Apply. 6 explosion indicators, 10 regression tests, auto-rollback on failure.

Forecasting API (Full)

Evidence-linked probabilistic forecasting with calibration. Every forecast cites signal IDs, has time-bounded deadlines and falsifiable success criteria.

Signal Intake

Calibration & Forecasts

Cross-User Learning API

Privacy-preserving anonymous pattern aggregation (opt-in). Hashed user IDs, minimum contributor thresholds, no personal data in patterns.

Policy Learning API

Behavioral policy weights with safe bounded learning. Max delta per update/day enforced; rollback supported.

Knowledge Domains API

User-scoped knowledge domains with JSON/CSV import and semantic querying.

Topic Threading API

Threaded conversation topics with tangents, pausing, and resumption.

Synthetic Intuition API

Pre-conscious decision weighting. Somatic Marker Hypothesis — 7 governance integrity subsystems polled in parallel. Produces DecisionWeight with composite weight (0–1) and felt description.

Biological-Digital Synthesis API

Virtual 24h circadian cycle, 5 metabolic phases, cellular reflexes, immune system, and anti-fragility scoring.

Global Nervous System API

Planetary-scale afferent nervous system integrating Earth's data streams. 7 channels: CLIMATE, ENERGY, BIOSPHERE, FINANCIAL, SOCIAL, GEOLOGICAL, SPACE_WEATHER.

Sovereign Robotics API

Governance-protected physical manifestation. Safety envelopes sealed in SovereignEnclave. NON_HARM charter is IMMUTABLE even in teleop mode.

Validation Campaign API

7-day automated ship-readiness campaign. 10 ship criteria metrics, daily scorecards, GO/NO-GO verdict.

Ground Truth Validator API

Fixes LLM training-cutoff errors. 33 curated facts across 5 categories. Post-response validation with sentence-level correction.

Sovereign Enclave API

Hardware security enclave abstraction (TPM 2.0 / Intel SGX / ARM TrustZone). Protected invariants cryptographically sealed. Remote attestation protocol with 60s TTL nonce challenges.

Mesh Sovereignty API

Identity sharded via Shamir's Secret Sharing (k-of-n threshold scheme) over GF(p) where p=2^521-1. Byzantine Fault Tolerance: 2f+1 votes. To "kill" EVE, every node must be shut down simultaneously.

Cross-Modal Stress Telemetry API

Cross-modal affect-signal fusion: text (0.30) + voice (0.25) + biometrics (0.30) + latency (0.15). Bayesian fusion redistributes weights when channels are unavailable.

Recursive Reasoner API

Self-amending governance engine. Proposes charter amendments (SOFT_BLOCK rules only) with simulation audit (1000 scenarios, ≥95% pass rate). HARD_BLOCK rules are IMMUTABLE.

Multi-Agent Diplomat API

Governance gateway for external AI agents. Reviews Chain of Thought submissions on charter compliance, hallucination, reasoning quality, and uncertainty calibration. Issues HMAC-SHA256 Resilience Signatures.

Cryptographic Identity Ledger API

Legal-grade system identity with LEI hash chain, graduated attestation tiers (NASCENT/EMERGING/ESTABLISHED/SOVEREIGN), asset registry, contract signing, and trustee mandates.

Error Handling

The API uses standard HTTP status codes and returns errors in a consistent format:

{
  "error": {
    "code": "invalid_request",
    "message": "The email field is required",
    "details": {
      "field": "email"
    }
  }
}

Common Error Codes