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.

Codex/ChatGPT multiple account load balancer & proxy with usage tracking, dashboard, and OpenCode-compatible endpoints

Soju06/codex-lb ↗

Soju06/codex-lb AI coding statistics: 26.5%

of 1,220 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate
26.5% agent · top contributor 83.3%31 contributors
individual

One person is driving this. The headline number reflects a single contributor's workflow, not a project-wide shift.

323
agent commits
897
other considered commits
31
agent contributors
83.3%
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

322assisted
1autonomous

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
2026-020%48
2026-030%99
2026-042%99
2026-050.6%161
2026-060.9%111
2026-0741.5%224
2026-0835%240
2026-0959.7%238

Peak month: 59.7%. Bars are scaled to that peak, not to 100%.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic31225.6%
Cursor AgentAnysphere60.5%
GitHub CopilotGitHub30.2%
OpenAI CodexOpenAI20.2%

Inspect the matches

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

Matched commit samples

  • f94dd509 on GitHubfix(utils): consume a shared future's exception when no waiter is left (#2Claude Code
  • 982125a3 on GitHubrevert(overflow): remove the subscription-exhaustion model-source fallbackClaude Code
  • 837964b9 on GitHubfeat(auth): add an OIDC identity provider with UI-configured, encrypted coClaude Code
  • 6d788664 on GitHubfix(proxy): keep source Responses SSE alive and normalized (#1854)Cursor Agent
  • 09dd9348 on GitHubfix(proxy): exclude previous_response_id from model-source routing (#1859)Cursor Agent
  • 254839f5 on GitHubfix(proxy): preserve explicit null embedding fields (#1836)Cursor Agent
  • c8fcc65a on GitHubfeat(proxy): add SOCKS4/SOCKS5 outbound proxy support via env vars (#1008)GitHub Copilot
  • a8cbefa8 on GitHubrefactor: upgrade error handling and replace print with standard logging iGitHub Copilot
  • 36a4e7cb on GitHubdocs: add comprehensive docstrings to select_account in logic.py (#350)GitHub Copilot
  • 7a0b6719 on GitHubfix(proxy): durably recover hard HTTP bridge operations (#1657)OpenAI Codex
  • 6b1cb47e on GitHubfix(chat): resolve parallel tool_call argument duplication via output_indeOpenAI Codex

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 window1,404
Merge commits removed−1
CI / dependency bots removed−183
Considered1,220

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 Soju06/codex-lb

Scan another repository