Review economics
HKUDS/LightRAG
Read from the 100 most recently merged pull requests ·
Pull requests with attributed agent authorship took only 0.3× the reviews submitted of the rest.
≥51%
Attributed
a floor
47
Attributed PRs
53
Other PRs
100
Read
Attributed — the latest 5
#3581 Rank isolated entities in get_popular_labels, stop dropping edges, unify the graph read contract
0 reviews · 1h to merge
#3580 refactor(tools): make tiktoken and spaCy cache downloads independent
0 reviews · 0h to merge
#3578 test(kg): non-interactive CLI for test_graph_storage.py, fix swallowed failures, drop redundant pg-smoke step
0 reviews · 1h to merge
#3575 Port graph storage contract tests to pg_smoke suite
0 reviews · 0h to merge
#3570 Reject unsafe upload filenames and use secure opener to prevent traversal/symlink attacks
4 reviews · 4h to merge
The rest — the latest 5
#3574 fix(rerank): validate results at query boundary
0 reviews · 6h to merge
#3573 fix(api): clean up startup dependency warnings (SVG rasterizer + smart_heading)
0 reviews · 0h to merge
#3572 fix(docker): install libcairo2 so native md SVG rasterization works
0 reviews · 0h to merge
#3568 fix(kg): address PGTableGraphStorage review findings — startup scans, read fan-out, docs, CI reach
0 reviews · 4h to merge
#3103 feat(kg): add PGTableGraphStorage — PostgreSQL-native graph backend (no AGE dependency)
18 reviews · 75.7d to merge
How this was measured
51% of merged commits carry agent attribution — a floor, not the share; tools that only complete code inline leave no commit trail, so the unattributed side includes AI-assisted work; 2 PRs came off an agent branch with no commit attribution, so the real share runs higher; reads 100 of 1506 merged PRs — the rest have not been analyzed yet.
Detected: Claude Code, OpenAI Codex.
“Attributed” means a commit carried an agent’s signature — a co-author trailer, an agent commit identity, or an agent bot account. Tools that only complete code inline leave no such mark, so the other column is “rest”, not “human-written”. Full method and its limits
The badge reads this repo’s current report, so it follows the number.
Tell me when this moves
We re-read HKUDS/LightRAG weekly and email only when the number changes materially.
Read your own repos
This one is public. For a private repo, run the same read locally through your own GitHub credentials — nothing is installed and nothing is sent to us.
npx @ambera/review-taxWant this continuously on HKUDS/LightRAG — each pull request paired to the task it came from, and the work graded from its review loop rather than its diff size? Claim this repo in Forge
Browse every repo people have read · Read a different repository