I Built KIRA: A Local-First AI Agent That Has to Prove Its Work

Описание и применение

KIRA — локальный агент, который подтверждает выполнение задач реальными результатами инструментов, обеспечивая прозрачность и безопасность.

Задача

AI assistants often claim to have completed tasks without verifiable evidence, making it hard to trust their actions.

Результат

KIRA demonstrates a reproducible, permission‑aware local AI agent that can be tested on real workflows, providing transparent evidence of task completion.

Рабочий процесс

• 1. User states a goal
• 2. Orchestrator V1 selects next tool
• 3. KIRA runs the tool with permission gates
• 4. Tool returns real result
• 5. Orchestrator evaluates evidence
• 6. Repeat until task genuinely complete

Доказательства

KIRA should not say a task is complete unless the runtime has evidence from the tool that performed it.

Источник

The installer downloads Orchestrator V1 into a local model directory.

Источник

KIRA keeps mutating file, command, IDE, MCP, and automation actions behind explicit permission gates.

Источник