swe-sweep

How many bugs can LMs find & fix in large codebases?

facebookresearch/swe-sweep

View on GitHub ↗

How many bugs can LMs find & fix in large codebases?

swe-sweep grew +35.0% in the last 30 days, ranks 421st of 872 in Evaluation & Guardrails, actively maintained, last commit 9 days ago.

Star trajectory

48 days tracked
2026-10-05247 stars today2026-11-22
Momentum

+35.0% in 30d, above the Evaluation & Guardrails median of +1.4%.

Maintenance

Last commit 9 days ago, fresher than the median of 15d.

How it compares in Evaluation & Guardrails

swe-sweep
247
stars
+35.0% 30d
commit 9 days ago
Evaluation & Guardrails median
235
stars
+1.4% 30d
commit ~15d ago
Leader · impeccable
114.7k
stars
+0.0% 30d
commit today

Ranked 421st of 872 by stars.

Ranked by shared tags, then keywords, then stars
  1. Benchmarking models as runtime Controllers for Loop Engineering

    +4 (+2.6%) in 30d · Committed 38d ago · Python · Apache-2.0

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  2. Give a coding agent a brief, not a chat, and it works on its own across sessions. Includes an example run: an…

    +4 (+1.0%) in 30d · Committed 34d ago · TypeScript · MIT

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  3. Bug Hunt Bench: 105 real bugs in two production repos, frontier coding models (GPT-6, Claude, Grok, Gemini…

    +65 (+42%) in 30d · Committed 2d ago · JavaScript

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  4. Taking the pi coding agent to Claude Code-level performance: 0.539 → 0.773 pass@1 on Terminal-Bench 2.1 with…

    +1 (+0.5%) in 30d · Committed 46d ago · TypeScript · MIT

    Shares Coding Agents & Assistants, Evaluation & Guardrails

  5. Formal verification for AI: a Lean 4 + Dafny proof checker that catches fake AI proofs (reward hacking), and…

    +60 (+2.1%) in 30d · Committed 30d ago · Python · MIT

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  6. Catches and repairs AI hallucinations before delivery: three quality gates, 31 built-in rules, a self-healing…

    -1 (-0.3%) in 30d · Committed today · Rust · Other license

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  7. A lightweight harness designed to be simple to use, efficient, universially compatible with any coding agent…

    +2 (+0.8%) in 30d · Committed 35d ago · Go

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  8. Playwright for coding agents. Test that your skills, MCP servers, and CLIs actually work when an agent uses…

    +6 (+3.9%) in 30d · Committed yesterday · Python · Apache-2.0

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  9. Behavioral eval lab (Quorum) for the superpowers project that drives real coding-agent CLIs (Claude, Codex…

    +14 (+10%) in 30d · Committed 32d ago · TypeScript

    Shares Evaluation & Guardrails, Coding Agents & Assistants

  10. Official Code for the paper "Doc2Harness: Benchmarking LLMs on Long-Horizon Harness Engineering of Coding…

    +4 (+3.5%) in 30d · Committed 26d ago · Python · Other license

    Shares Evaluation & Guardrails, Coding Agents & Assistants

Language
Python
License
Other
Contributors
4
Forks
31
Watchers
0
Open issues
12
Releases
0
Created
2026-10-01
Stars this week
+7

You've seen where swe-sweep stands today.

Get the moves before everyone else: the 5 Evaluation & Guardrails repos climbing fastest, every Tuesday.

Free weekly digest. Double opt-in, unsubscribe anytime.

Maintain swe-sweep?

Claim this page to add your own call-to-action.

Claim this page