Stop Claude Code from doing irreversible damage. Policy-gated execution + receipts so you can ship agents without sweating production.

Deterministic agent runtime with explicit plans, typed tools, permissions, and replay. Implements Google DeepMind's Intelligent AI Delegation framework. Named after ELP's Karn Evil 9.

Constitutional Verification Model (CVM), a layered framework for reasoning about how decentralized systems achieve legitimate rule authority and independently verifiable execution through invariant correctness, deterministic execution, and replayable history.

Execution OS with human approval gates and pressure-driven task control.

A kernel-userland protocol enforcing information-theoretic bounds on AI adaptivity leakage, benchmark gaming, and capability spillover.

Offline-first LLM orchestration CLI for planning and executing human-approved tasks via validated git patches, with reproducible runs and full auditability.

HELM OSS — Open-source core of the HELM Autonomous OS. Policy enforcement, kernel runtime, proof graphs, and audit infrastructure.

AIΩN – A formal framework for deterministic, history-native computation.

R-LAM is a reproducibility-constrained execution framework for Large Action Models in scientific workflow automation. It enables adaptive, agent-driven workflow execution while enforcing strict guarantees on auditability, determinism, and replayability.

Example skills, manifests, and reference projects for building on Inactu.

Deterministic execution boundary for AI systems enforcing signed approvals, replay protection, and cryptographic receipts.

Built to answer: How do you execute millions of smart contract transactions — concurrently — without breaking determinism?

Secure execution substrate for immutable agent skills with explicit capabilities, cryptographic provenance, and auditable deterministic runs.

AINL helps turn AI from "a smart conversation" into "a structured worker." It is designed for teams building AI workflows that need multiple steps, state and memory, tool use, repeatable execution, validation and control, and lower dependence on long prompt loops. AINL is a compact, graph-canonical, AI-native programming system for building deter

MCP server for deterministic task execution via Domain Profile Agents + Task Templates. Provides planning/orchestration layer for CLI AI assistants (Warp, etc.)

The Universal Goal Execution Model. A deterministic computational substrate for autonomous agents and post-procedural software. Stop programming execution. Start declaring intent.

A strict, auditable execution engine for deterministic outbound communication (email + voice). Same inputs, same outputs — every time.