Pulse
2026-09-18-v3.0
Dataset
2026-10-27-catalog-ai-v2
Ruleset
b2e8889c66f7
Scanned
2026-10-27
Window
February–September 15, 2026
Catalog tracking observation. This repository is measured with the same fixed window and ruleset, but it is not part of the published benchmark or its aggregate claims.

SkyRL: A Modular Full-stack RL Library for LLMs

NovaSky-AI/SkyRL ↗

NovaSky-AI/SkyRL AI coding statistics: 37.6%

of 647 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate
37.6% agent · top contributor 24.3%33 contributors
org-wide

Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.

243
agent commits
404
other considered commits
33
agent contributors
24.3%
largest contributor

What shapes this result

Mode, timing, and detected signatures show whether the result reflects broad adoption or one concentrated workflow.

Assisted or autonomous

243assisted
0autonomous

Assisted — a developer’s identity authored the commit and the agent added itself as co-author. Autonomous — a bot identity authored the commit without a human identity attached.

Agent share by month

Agent-attributed commit share by month
MonthAgent shareConsidered commits
2026-0232.3%158
2026-0335.3%102
2026-0420.2%89
2026-0556.7%60
2026-0634.6%78
2026-0737.3%67
2026-0852.9%51
2026-0959.5%42

Peak month: 59.5%. Bars are scaled to that peak, not to 100%.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic19830.6%
Gemini CLIGoogle457%
DevinCognition162.5%
Cursor AgentAnysphere111.7%

Inspect the matches

Public commit samples let you inspect the signatures behind the aggregate result.

Matched commit samples

  • 58891b2f on GitHub[megatron] 5/n towards Kimi K2.6: Kimi K2.5-family (KimiK25ForConditionalGClaude Code
  • c0241450 on GitHub[megatron] 11/n towards Kimi K2.6: file-backed offload of frozen masters sClaude Code
  • 8f9a3e68 on GitHub[fix][tinker] 14/n towards Kimi K2.6: sleep colocated engines around TinkeClaude Code
  • 8a151169 on GitHub[chore] Upgrade `flashinfer` to 0.6.14 (#2135)Gemini CLI
  • 495ecb7e on GitHub[models] add nemotron 3.5 lightning to gpu CI and add example script (#213Gemini CLI
  • a494186b on GitHub[deps][megatron] bump megatron-bridge to 0.7.0 and megatron-core to 0.20.0Gemini CLI
  • 7f65395e on GitHub[tinker] Support prompt_logprobs in SkyRLTrainBackend sample() path (#1461Devin
  • 7f5eba15 on GitHub[feat][R3] Enable R3 with new inference (#1428)Devin
  • c20a2930 on GitHub[feat][train] Support prompt_logprobs in /sample in the new inference stacDevin
  • 58891b2f on GitHub[megatron] 5/n towards Kimi K2.6: Kimi K2.5-family (KimiK25ForConditionalGCursor Agent
  • c0241450 on GitHub[megatron] 11/n towards Kimi K2.6: file-backed offload of frozen masters sCursor Agent
  • 8f9a3e68 on GitHub[fix][tinker] 14/n towards Kimi K2.6: sleep colocated engines around TinkeCursor Agent

What entered the denominator

The result begins with commit history, removes merges and routine automation, and reports literal agent signatures as a lower bound.

Exclusion ledger

Commits in window647
Merge commits removed−0
CI / dependency bots removed−0
Considered647

Bots are excluded from the denominator entirely. Agent attribution requires a literal supported signature; no code-style inference is used.

Reproduce this observation

Run the same detector locally.

Deterministic signature matching; no code-style inference. Squash merges can remove trailers, so this result remains a measured floor.

npx qmmit-cli NovaSky-AI/SkyRL

Scan another repository