← the index
milvus-io/milvus ↗
36.5%
of 1,598 commits were committed by an AI agent — a floor, not an estimate
36.5% agent · top contributor 17%41 contributors
org-wide

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

584
agent commits
1,014
human commits
41
agent contributors
17%
largest contributor

Assisted or autonomous

584assisted
0autonomous

Assisted — a developer’s identity authored the commit and the agent added itself as co-author. The agent ran git commit during a working session, with a person there. Autonomous — a bot identity is the author, no human in the loop. Across the whole index, 93.7% of agent commits are assisted.

Agent share by month

Feb26
Mar26
Apr26
May26
Jun26
Jul26
Aug26
Sep26

Peak month: 53.8%. Bars are scaled to that peak, not to 100%.

Which agents

Claude CodeAnthropic57736.1%
OpenAI CodexOpenAI110.7%
Cursor AgentAnysphere20.1%

Verify it yourself

Sample commits that matched. Every one links to GitHub — check the trailers.

  • 310377affix: fail index build fast on malformed user JSON instead of retrying foreClaude Code
  • 39cca203enhance: make arrow IO thread pool coefficient 0 a fixed Milvus default (#Claude Code
  • 10a6bd99enhance: make configuration projections safe by default (#52579)Claude Code
  • 240b9c0afeat: publish StorageV3 index artifacts through segment manifests (#53048)OpenAI Codex
  • 82ccbec2fix: validate V1 binlog reader completion (#53255) (#53256)OpenAI Codex
  • bf65725bfix: skip file resource sync before first add (#52707)OpenAI Codex
  • ea0c5332enhance: make tombstone sweeper non-blocking and concurrency-safe (#52636)Cursor Agent
  • 490a5c36fix: fix bloom filter leak when worker node crashes (#47450)Cursor Agent

What was excluded

Commits in window1,599
Merge commits removed−0
CI / dependency bots removed−1
Considered1,598

Bots are excluded from the denominator entirely. Counting dependabot as “AI-written code” is the fastest way to produce a meaningless number.

How this was measured

Measured over February–August 2026, cloned directly. Detection is deterministic string matching against signatures agents write about themselves — no heuristics, no style analysis. Merge commits and CI bots are excluded from the denominator. The figure is a floor: squash merges strip trailers, so real usage is higher. Contributor counts are aggregate — no individual is named.

Run it yourself: npx qmmit-cli milvus-io/milvus

Scan another repository

Try