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.

⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs. Built-in failover, load balancing, cost control & end-to-end tracing.

01Result

looplj/axonhub ↗

looplj/axonhub AI coding statistics: 4.1%

of 800 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate
4.1% agent · top contributor 21.2%21 contributors
org-wide

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

33
agent commits
767
other considered commits
21
agent contributors
21.2%
largest contributor

02Drivers

What shapes this result

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

Assisted or autonomous

33assisted
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
2026-021.7%118
2026-037.3%96
2026-042.8%142
2026-0513.2%68
2026-064.3%93
2026-071.2%83
2026-082.8%107
2026-093.2%93

Peak month: 13.2%. Bars are scaled to that peak, not to 100%.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic172.1%
Gemini CLIGoogle81%
GitHub CopilotGitHub30.4%
DevinCognition30.4%
Cursor AgentAnysphere20.3%

03Evidence

Inspect the matches

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

Matched commit samples

  • fd4158ef on GitHubfix(openai): merge multiple system messages for strict upstreams (#2384)Claude Code
  • d46fbcf4 on GitHubfix(images): accept application/json in passthrough mode for /images/editsClaude Code
  • 1a9f4560 on GitHubfix(openai): keep message content valid for strict upstreams (#2137)Claude Code
  • 5e020e51 on GitHubfeat: add Qiniu provider support (#1612)Gemini CLI
  • fc624585 on GitHubfeat(oidc): Initial OpenID Connect Support (#1091)Gemini CLI
  • 0f063a42 on GitHubopt(responses): fix function parameters missing properties, close #1032 (#Gemini CLI
  • fe74b95f on GitHubfix(copilot): send Grok models through Responses API (#2250)GitHub Copilot
  • 1eff8fb9 on GitHubfeat: realtime chunks preview and response preview (#1342)GitHub Copilot
  • cae8a908 on GitHubfeat(config): auto generate schema for config (#1100)GitHub Copilot
  • 23e553d6 on GitHubfeat: introduce Atlas Cloud as sponsor (#1693)Devin
  • e2d96b3d on GitHubfix: anthropic adaptive thinking effort did not pass to other outbound (#1Devin
  • c3a3aed7 on GitHubopt(llm): thinking signature encode with channel footprint, close #1035 (#Devin
  • ad5b5eec on GitHubfeat(channel): global auto-disable rules and bulk apply (#2453)Cursor Agent
  • afaf90f9 on GitHubfix(frontend): guard execution channel query by permission (#1770)Cursor Agent

04Method

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 window883
Merge commits removed−0
CI / dependency bots removed−83
Considered800

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

Continue exploring

Claude Code commitsBrowse repositories with detected Claude Code signatures →GitHub Copilot commitsBrowse repositories with detected GitHub Copilot signatures →Cursor Agent commitsBrowse repositories with detected Cursor Agent signatures →Devin commitsBrowse repositories with detected Devin signatures →Gemini CLI commitsBrowse repositories with detected Gemini CLI signatures →Go AI coding statisticsCompare evidence across Go repositories →GitHub AI commit analyzerInspect another repository with the same evidence model →
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 looplj/axonhub

Scan another repository

Try