SWE-agent

tom@tomrochette.com (Tom Rochette) — Wed, 28 Oct 2026 00:00:00 +0000

SWE-agent is the Princeton and Stanford research harness that introduced the agent-computer interface (ACI) idea, carried the model into a real repository through a constrained shell tool set, and became the reference agent of the early SWE-bench era.

What it is
#

An open-source (MIT) Python terminal harness from the SWE-bench team, published with a NeurIPS 2024 paper arguing that interface design, not just model quality, decides agent performance. The agent works inside a Docker-containerized copy of a repository through a yaml-governed set of tool bundles (an edit-lint flow, search, and a submit gate), and every run leaves a complete .traj trajectory file. It is built and maintained by researchers from Princeton University and Stanford University, sits beside their SWE-bench, SWE-ReX, and SWE-smith projects, and carries an EnIGMA mode for offensive-security capture-the-flag work pinned to the older v0.7 line.

Status
#

Maintenance-only by its own declaration, and still the category’s most cited ancestor: 20,663 stars on a repository created April 2, 2024, with the default branch pushed October 26, 2026 (GitHub API, as of 2026-10-28). The newest tagged release is v1.1.0 (May 22, 2025, a training-trajectories drop), and no release has followed in seventeen months, while the docs site and README now carry a standing banner: most development effort moved to mini-swe-agent, which “has superseded SWE-agent”, and “SWE-agent is now in maintenance-only mode”. The April 2, 2024 launch thread drew 307 points and 143 comments on Hacker News, and the paper’s 12.5 percent SWE-bench pass@1 was the open-source state of the art at publication.

Strengths
#

The ACI thesis is the field’s clearest statement that harness design is a first-class variable, and the paper’s ablations still anchor the argument.
Everything is yaml: model, templates, history processors, demonstration trajectories, and tool bundles, which makes run configurations reproducible and shareable.
Model routing through litellm covers essentially every provider including local Ollama-style endpoints, with a custom cost registry for unpriced models.
Batch mode (run-batch) with per-thread API-key rotation was built for benchmark evaluation, and .traj outputs make every run auditable.
Prompt-caching support for Claude via a cache_control history processor shows unusual attention to inference cost for an academic tool.

Cautions
#

The project recommends its own replacement: the supersession banner names mini-swe-agent as the going-forward choice, and maintenance-only means no release since May 2025.
The SWE-bench numbers it rode on are themselves contested: the SWE-Bench+ analysis found solution leakage in 32.67 percent of SWE-Agent+GPT-4’s successful patches and weak tests in another 31.08 percent, dropping its measured resolution rate from 12.47 to 3.97 percent.
The 1.0 rewrite broke the 0.7 configuration surface, and EnIGMA (the security mode) still documents the old version as its supported line.
No subagents, no MCP, no editor integration: the loop is one model over one shell, by design.

Pricing
#

Free and open source under MIT; pricing does not apply. The cost of running it is your model API spend, which the litellm registry tracks per run.

Compared to
#

Mini-SWE-agent is the same team’s successor, a deliberately tiny loop with matching SWE-bench performance, and the project’s own recommendation for anything new. OpenHands is the platform-scale research agent, with a self-hostable UI and cloud, where SWE-agent stays a scriptable command-line research instrument. Claude Code and the other commercial members of this category are production harnesses; SWE-agent is the instrument the category’s evaluation culture grew up around.

Bottom line
#

Recommended for research on agent-harness design and for reproducible SWE-bench batch runs where the .traj records and yaml configs are the point. Not for daily production coding, where the project itself points you to mini-swe-agent or a commercial harness.

Changes
#

2026-10-28 - Created from the cross-category handoff resolving SWE-agent/SWE-agent into the harnesses category.

References
#

https://api.github.com/repos/SWE-agent/SWE-agent - repository metadata: stars, license, push date, and creation date (as of 2026-10-28)
https://github.com/SWE-agent/SWE-agent - README: the supersession banner, feature claims, EnIGMA, and the sibling projects
https://swe-agent.com/latest/ - the documentation home: the maintenance-only banner, news, and doc structure
https://swe-agent.com/latest/config/models/ - litellm model routing, local models, Claude prompt caching, and key rotation
https://swe-agent.com/latest/faq/ - local model support, custom tools, multimodal configuration, and .traj outputs
https://swe-agent.com/latest/usage/adding_custom_tools/ - the tool-bundle structure behind the tools row
https://arxiv.org/abs/2405.15793 - the NeurIPS 2024 paper: the ACI thesis and the 12.5 percent SWE-bench result
https://arxiv.org/abs/2410.06992 - SWE-Bench+: the critical analysis of SWE-Agent+GPT-4 resolution rates, leakage, and weak tests
https://hn.algolia.com/api/v1/items/39907468 - the 307-point April 2, 2024 launch thread

swebench on tomrochette.com

SWE-agent

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#

swebench on tomrochette.com

SWE-agent

What it is #

Status #

Strengths #

Cautions #

Pricing #

Compared to #

Bottom line #

Changes #

See also #

References #

What it is
#

Status
#

Strengths
#

Cautions
#

Pricing
#

Compared to
#

Bottom line
#

Changes
#

See also
#

References
#