---
format: "aidr-story-markdown/v1"
id: "a9c380b2d278d8eae5077682073f7bc46d610a6f5a5f84a069cd097b7372e2ba"
canonical_url: "https://aidr.today/a9c380b2?lang=en"
title: "AgentSWE: Can Coding Agents Build the Agent You Actually Want?"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-13T00:00:00.000Z"
category: "Open Source"
topics: ["agent","open-source"]
source_urls: ["https://huggingface.co/papers/2610.13361"]
summary: "AgentSWE is a benchmark for agent software engineering that spans three stages of the agent development lifecycle: creation, editing, and optimization. It features 25 tasks graded against requirement-derived rubrics or source benchmarks."
---

# AgentSWE: Can Coding Agents Build the Agent You Actually Want?

> [Open the canonical story](<https://aidr.today/a9c380b2?lang=en>)

**Published:** 2026-10-13T00:00:00.000Z
**Category:** Open Source
**Topics:** agent, open\-source

## Summary

AgentSWE is a benchmark for agent software engineering that spans three stages of the agent development lifecycle: creation, editing, and optimization\. It features 25 tasks graded against requirement\-derived rubrics or source benchmarks\.

## Sources

- [Story source](<https://huggingface.co/papers/2610.13361>)