Split large PR diffs into reviewable sub-patches via LLM-assigned groups.

reviewdeck is the product name for this first public release.

Run it ad hoc:

npx reviewdeck@^0.2.0 index pr.diff

Or install it into another repo:

npm install -D reviewdeck

The repo exposes a standard skill at skills/reviewdeck, so external users can install it with npx skills add:

npx skills add neutree-ai/reviewdeck --skill reviewdeck

Useful variants:

# list skills in the repo first
npx skills add neutree-ai/reviewdeck --list

# install globally instead of into the current project
npx skills add neutree-ai/reviewdeck --skill reviewdeck --global

gh pr diff 123 > pr.diff
npx reviewdeck@^0.2.0 index pr.diff > pr.index.txt

Have the agent use the reviewdeck skill to turn the indexed changes into split metadata JSON, then:

cat split.json | npx reviewdeck@^0.2.0 split pr.diff - -o output/
npx reviewdeck@^0.2.0 render output/

If the goal is PR review, render is the normal next step after split; split only proves the sub-patches are valid and ordered. render is meant to produce a human review UI or artifact, not to replace the human review with automatic findings. When generating split metadata, the skill can follow lightweight review patterns such as deps-first or tests/docs-first; if the user has no clear preference, deps-first is the default. The split metadata descriptions are most useful when they explain why the changes are grouped together or what the reviewer should verify in that step. The split metadata can also include optional group-level draftComments, which render shows inline as agent co-review drafts for the human reviewer to accept or reject. When render submits, it prints a JSON object with final comments plus draftComments status so an agent can tell which drafts were accepted before deciding whether to post them back to a PR.

vp install
vp test            # run tests
vp check           # format + lint + type check
npm run lint       # full quality pipeline
npm run pack:dry-run