{"id":47270797,"title":"Show HN: SafeAgent – exactly-once execution guard for AI agents","points":2,"user":"Lions2026","time":1772770487,"time_ago":"16 hours ago","type":"link","content":"

LLM agents and tool-calling systems frequently retry or replay actions.

This becomes dangerous when the action is irreversible:\n- sending emails\n- opening tickets\n- executing trades\n- triggering payouts

SafeAgent is a small Python library that enforces exactly-once execution for agent actions using request-ID deduplication.

Example:

from settlement.settlement_requests import SettlementRequestRegistry

registry = SettlementRequestRegistry()

def send_email():\n print("EMAIL SENT")

registry.execute(\n request_id="email:user123:invoice",\n action="send_email",\n payload={"to": "user123@example.com"},\n execute_fn=send_email\n)

If the same request_id is replayed, the action does not run again.

Instead the original execution receipt is returned.

pip install safeagent-exec-guard

GitHub:\nhttps://github.com/azender1/SafeAgent

PyPI:\nhttps://pypi.org/project/safeagent-exec-guard/","url":"item?id=47270797","comments":[{"id":47271287,"level":0,"user":"stratifyintel","time":1772775443,"time_ago":"15 hours ago","content":"

This becomes a real issue once agents start interacting with external systems. Retries are easy to miss during development, but once you add tool calls + network failures the agent loop can replay same action multiple times.

Idempotency/request IDs end up becoming pretty important for anything that touches emails, payments, tickets etc","comments":[]}],"comments_count":1}