Temper

Deploy formally verified agent operating systems with Cedar authorization.

46|11|Updated Feb 9, 2026
One-click install
npx skills add https://github.com/nerdsane/temper --skill temper
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Temper
Source: https://github.com/nerdsane/temper/tree/main/skills/temper-openclaw
Command: npx skills add https://github.com/nerdsane/temper --skill temper

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

Temper provides a framework for agents to define, verify, and deploy their own operating systems, managing state, authorization, and evolution through formally verified specifications.

Core Features & Use Cases

  • Declarative Specs: Agents define state machines, data models, and authorization policies in TOML.
  • Formal Verification: Specs undergo rigorous L0-L3 verification (SMT, model checking, simulation, property tests) before deployment.
  • Runtime API: Deploys verified specs as a live OData API for agent interaction.
  • Policy Mediation: All actions are mediated by Cedar policies, enforcing default-deny authorization.
  • Durable Execution: State is persisted using Turso (libSQL), surviving restarts.
  • UI Generation: Provides a design system and proxy for building dynamic UIs that reflect live state.
  • OpenClaw Integration: Seamlessly connects agents via SSE and signal files for real-time state updates.
  • Use Case: An agent needs long-term memory. It defines a Knowledge.ioa.toml spec for storing and retrieving entries. Temper verifies, deploys it as an API, and the agent uses this API to manage its knowledge base, with all access controlled by Cedar policies.

Quick Start

Use the temper skill to start the Temper server locally with default settings.

Frequently Asked Questions about Temper

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do agents define and verify their own operating systems using declarative specifications?

Agents define operating systems using declarative specifications by writing state machines, data models, and authorization policies in TOML, which then undergo rigorous L0-L3 formal verification including SMT, model checking, and property tests before deployment.

What is formal verification for agent state machines and when is it needed?

Formal verification for agent state machines is a rigorous L0-L3 validation process using SMT, model checking, and simulation to ensure specs are safe before deployment. It is needed when agents require formally verified tools managing persistent state.

How do I manage agent authorization policies using Cedar?

You manage agent authorization policies using Cedar by enforcing default-deny mediation across all actions. Policies are derived directly from behavioral specs, ensuring all runtime API interactions are strictly controlled.

Does Temper support durable state persistence for agent operating systems?

Yes, Temper supports durable state persistence for agent operating systems by using Turso with libSQL. This ensures that all state data survives restarts and remains accessible during runtime API operations.

How do agents interact with deployed state machine specs via OData API?

Agents interact with deployed state machine specs via a live OData API. Once declarative specs pass formal verification, they are deployed as this API, enabling agents to manage state and execute controlled actions.

Can I integrate real-time UI updates and SSE with agent state machines?

Yes, you can integrate real-time UI updates and SSE with agent state machines. The framework provides a design system and proxy for dynamic UIs, while OpenClaw integration uses SSE and signal files to reflect live state changes.