SafeAgent – exactly-once execution guard for AI agent side effects

Description

LLM agents retry tool calls constantly. Retries can happen because of: - model loops - HTTP timeouts - queue retries - orchestration restarts If the tool triggers something irreversible you can end up with duplicate side effects: retry → duplicate payment retry → duplicate email retry → duplicate ticket retry → duplicate trade SafeAgent is a small Python guard that sits between the agent decision and the side effect. Pattern: agent decision → deterministic request_id generate