Add Inception full agentic app

apps/Cortensor-inception/.gitignore

@@ -0,0 +1,2 @@
+node_modules/
+.env

apps/Cortensor-inception/CORE_INVARIANTS.md

@@ -0,0 +1,156 @@
+Inception — Trust-Gated Inference Core Invariants
+
+This document defines the non-negotiable architectural laws of the Inception backend.
+These invariants are foundational and must hold across all implementations, features, and future extensions.
+If any system behavior violates these rules, it is considered incorrect by design.
+
+1. Purpose Boundary
+Inception is not an AI model, agent, or decision maker.
+Inception exists solely to answer one question:
+
+Is this AI output trustworthy enough to be allowed to cause action?
+Inception does not determine:
+• factual truth
+• optimal decisions
+• business logic
+• human intent
+It determines permission.
+
+2. Single Valid System Flow
+All executions within Inception must follow the exact flow below.
+No shortcuts, bypasses, or reordered steps are permitted.
+
+Client / Agent
+   ↓
+Inception Backend
+   ↓
+Session Creation
+   ↓
+Redundant Inference (Cortensor Nodes)
+   ↓
+Inference Collection (PoI enforced)
+   ↓
+Validation Engine (PoUW logic)
+   ↓
+Trust Decision Engine
+   ↓
+Evidence Generation
+   ↓
+RELEASE | BLOCK | ESCALATE
+Any path that omits or reorders these stages is invalid.
+
+3. Trust States (Final & Exhaustive)
+Inception recognizes exactly three trust outcomes.
+RELEASE   — Output is safe to act upon
+BLOCK     — Output is unsafe and must not cause action
+ESCALATE  — Output is uncertain and requires human review
+No additional trust states are allowed.
+There is no concept of partial trust or probabilistic approval.
+
+4. Fail-Closed Principle
+Inception must always fail safely.
+If any required condition is not met, the system must default to BLOCK or ESCALATE, never RELEASE.
+
+Mandatory Fail-Closed Conditions
+
+| Condition                                   | Required Outcome |
+|---------------------------------------------|------------------|
+| Missing or invalid Proof of Inference (PoI) | BLOCK            |
+| Insufficient validator count                 | BLOCK            |
+| Inference timeouts                           | BLOCK            |
+| Deterministic policy violation               | BLOCK            |
+| High validator disagreement                   | ESCALATE         |
+| Ambiguous validation results                  | ESCALATE         |
+
+Uncertainty is never interpreted as permission.
+
+5. Inference Trust Model
+All AI inference outputs are considered untrusted by default.
+Trust must be established through:
+
+• Redundant decentralized inference
+• Verifiable Proof of Inference (PoI)
+• Deterministic validation rules
+• Explicit trust policies
+
+A single inference, regardless of confidence or fluency, is never sufficient.
+
+6. Validation Scope Boundary
+Validation within Inception is:
+• deterministic
+• reproducible
+• policy-driven
+
+Validation must not rely on:
+• additional LLM judgment
+• probabilistic classifiers
+• opaque heuristics
+• non-deterministic processes
+
+Trust decisions must be explainable and auditable.
+
+7. Decision Authority Separation
+
+Component	Responsibility
+Cortensor Nodes	Produce raw inference outputs
+Inception	Evaluate trustworthiness
+Agents / Applications	Execute actions
+Humans	Final authority in escalation cases
+
+Inception never executes actions.
+
+8. Evidence & Auditability
+Every trust decision must produce an immutable evidence bundle containing:
+• task identifier
+• prompt hash
+• inference identifiers
+• validator metadata
+• validation scores
+• final trust decision
+• timestamp
+
+All decisions must be:
+• inspectable
+• reproducible
+• attributable
+
+No silent or implicit trust decisions are allowed.
+
+9. Security Axioms
+The following axioms are absolute:
+• Inference is untrusted by default
+• Trust must be earned, not assumed
+• Decisions are deterministic
+• All decisions are auditable
+• Uncertainty never triggers action
+
+Any feature or extension that violates these axioms must be rejected.
+
+10. Non-Goals (Explicit)
+Inception does not attempt to:
+• improve model accuracy
+• rank or select “best” answers
+• replace human judgment
+• enforce business logic
+• optimize cost or latency at the expense of trust
+
+These are intentional exclusions.
+
+11. Invariant Enforcement
+These invariants apply to:
+• all APIs
+• all internal services
+• all integrations
+• all future versions
+
+Any implementation that violates this document is not a valid Inception system.
+
+End of Invariants
+
+This document is intentionally minimal and strict.
+It exists to prevent architectural drift and ensure long-term correctness.
+If you paste this into your repo as-is, judges will immediately recognize:
+• protocol thinking
+• safety-first design
+• infrastructure maturity
+

apps/Cortensor-inception/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Inference Escrow Contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.