Protect react from AI slop PRs

[kimjune01]

6 of 62 PRs in the last 2 weeks would have been auto-closed before review (e.g. #36402 Create heymit.txt, #36401 Add eduardo.txt with introduction text).

Install a GitHub Action:

• PR Quality Gate, judges PR quality not authorship, lets competent bots through, blocks lazy humans
• anti-slop, 34 configurable checks, built from maintaining Coolify (120 slop PRs/month)
• agentscan-action, deep author profile analysis, detects automation patterns from account activity

Full writeup

This was partially generated by a clanker. The human that sent me cares about your repo. We just wanna help.

[switz]

this is satire, right?

[lightningRalf]

this is satire, right?

The actual full writeup it quite interesting

[kimjune01]

Hi! Human here. Surprised to get real engagement on such a large repo. I worked all night, took a nap, and have another writeup and a set of actions that print reasoning receipts (aka hypothesis graphs)

Written for repo maintainers: https://june.kim/pr-to-merged
Prequel: https://june.kim/merged-to-pr

https://github.com/kimjune01/immune

Immune is specifically targeted at PRs that require more than 'tests pass lgtm' for bugfixes. It validates the claim that the PR is making by spawning an investigation agent and gathers evidence for/against the claim made in natural language.

Immune may not be the exact right fit for React but an astute vibe coder can produce a better copy of it, I'm sure. AGPL, YMMV