idempotency

A detailed engineering walkthrough on making action-taking AI agents production-ready, covering failure modes like duplicate refunds, prompt injection, and runaway tool calls, with a full FastAPI + LangGraph architecture including idempotency, spend caps, human-in-the-loop interrupts, and a policy engine.

Pi officially published a tutorial on building a multi-user Agent on Cloudflare Durable Objects: a single Durable Object maps to one session, state is stored in built-in SQLite, and it supports multiple users sharing a link to chat collaboratively, as well as scheduled wake-up tasks. The tutorial also demonstrates an isolated design for session retrieval, idempotent request handling, and a mechanism for sharing a single alarm across multiple tasks.

LoopX addresses the hard engineering problem of whether a remote side has executed after a tool call timeout in long-horizon agents. It achieves idempotent recovery through stable operation/Job identities, reading back external facts before deciding how to recover, and compositional verification — and has open-sourced this local-first control plane for long-horizon agents.

作者分享了一次因重试导致重复扣费的线上事故，并介绍其正在构建的 FetchSandbox 工具——通过确定性故障注入检查工作流重试时的幂等性是否真正生效。

An empirical study crashing four production AI agent codebases (Cua, Skyvern, Corsair, Inconvo) mid-action and measuring duplicate side effects on retry, finding a shared root cause: none of them records which actions already completed, so retries re-execute work from an empty history.

Armin Ronacher details how Pi Durable enables agent processes to resume from the point of interruption after a crash: persistence is achieved through an "effect sandwich" (intent, execution, result storage) rather than event sourcing, combined with a versioned document store to manage agent application state, supporting deployment to platforms such as Vercel, Cloudflare DO, and E2B.

A practitioner asks how agent frameworks should handle write-tool timeouts, weighing idempotency keys, read-back verification, and human escalation when downstream APIs lack idempotency support.

ArrivalBench introduces an evaluation method that re-executes agent-generated data pipelines under adversarial replay schedules (late, duplicated, out-of-order records) and finds that 7-79% of pipelines certified by single-execution benchmarks are silently wrong when re-run.

A developer updates an open-source FastAPI/Pydantic AI-agent policy-boundary starter with resource-scoped idempotency, audit logging for ALLOW decisions, and independent validation signals, and asks the community what production control to add next.

UndoBench introduces a benchmark that decouples task competence from fault-recovery capability in tool-using AI agents across 36 workflows and 36 fault scenarios in 8 enterprise domains, finding nominal task completion of 83.54% masks conditional recovery success of just 46.72%, with naive retry producing duplicate side effects in 53.33% of trials.

The article explores the distinction between approval and completion states in agent workflows through a simulation test, highlighting the importance of idempotency and proper handling of retries to avoid duplicate actions.

The article discusses a critical automation failure mode where actions succeed but responses time out, leading to duplicates, and advocates for using stable operation IDs and state checks to improve agent evaluation robustness.

This article discusses handling provider outages in AI agent systems by implementing failover, model fallback, circuit breakers, and idempotent retries to prevent pipeline stoppages and duplicate actions.

The author is building Synathic, an open-source verification layer for AI-agent tool calls, and seeks feedback on whether timestamp-based checks suffice for proving causation in database writes, recommending operation IDs for attribution.

AIPass releases versions 2.7.20 and 2.7.21, introducing significant updates to agent memory management with a fleet memory push and passports that ship with the repository, ensuring identities travel with the repo while live memories stay untracked.

This article discusses the challenges faced while building a code formatter for Gren, including issues with the compiler's parser omitting comments and punctuation, ensuring idempotent formatting, and testing without a formal specification.

This paper reports a failure study of a production agentic software-delivery platform and introduces reliability primitives for non-idempotent agent delegation, focusing on identity adequacy and evidence adequacy to improve system robustness.

A developer describes how an AI agent wrote a Stripe handler that double-provisioned customers on webhook retries, and how using FetchSandbox MCP to simulate retries helped catch and fix the idempotency bug.

The author shares lessons from putting an AI agent with write access into production, explaining that retry logic inside the agent loop fails when the process dies. They advocate for treating side-effectful tool calls as durable background jobs with idempotency keys.

This article explains how using Postgres transactions for workflow state co-located with application data eliminates idempotency and atomicity problems in distributed workflows, providing exactly-once execution.