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.

Kernel source tree for Raspberry Pi-provided kernel builds. Issues unrelated to the linux kernel should be posted on the community forum at https://forums.raspberrypi.com/

raspberrypi/linux ↗

raspberrypi/linux AI coding statistics: 0%

of 13,873 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

No signed agent commits found

No agent signatures were found in the last 13,873 considered commits. That is a real result, not a failure to measure.

It does not mean no AI was used here. A developer can use an agent and commit by hand without leaving a detectable git signature.

2
agent commits
13,871
other considered commits
2
agent contributors
50%
largest contributor

What shapes this result

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

No agent signatures were detected, so there are no agent, mode, or timing drivers to break down.

Which agents

AgentVendorCommitsShare
Claude CodeAnthropic20%

Inspect the matches

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

Matched commit samples

  • c49f04e8 on GitHubinet: frags: strip GSO state from fragments before reassemblyClaude Code
  • c9dadb31 on GitHubsctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALLClaude Code

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 window13,897
Merge commits removed−24
CI / dependency bots removed−0
Considered13,873

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 raspberrypi/linux

Scan another repository