RFC 406i: The Hilarious Hammer We Need to Smash AI Slop in Open Source
# RFC 406i: The Hilarious Hammer We Need to Smash AI Slop in Open Source
Open-source maintainers are drowning in a sea of AI-generated slop, and RFC 406i is the life raft we didn't know we needed—hilarious, ruthless, and spot-on. This fake RFC, dubbed "The Rejection of Artificially Generated Slop (RAGS)," mimics the dry precision of real internet standards but packs a punch: it outlines how to detect, diagnose, and obliterate those half-baked pull requests, issues, and comments spewed by Copilot or Claude. Forget polite rejections; this bad boy calls out your "bad prompt engineering" and threatens to downgrade your repo perms to WISHFUL_THINKING.
<> "Upon lexical and structural analysis of your submission, we have concluded that your prompt engineering is bad, and you should consequently feel bad."/>
Chef's kiss. That's the kind of unfiltered truth that Hacker News ate up, racking up 186 points and 63 comments of pure maintainer catharsis. Users are cheering lines like blacklisting MAC addresses for repeat offenders or routing PRs through a cyan-less dot-matrix printer. One commenter nailed it: "The keywords 'MUST NOT' are interpreted exactly as how much we do not want to review your generated submission." It's satire, sure, but it exposes a brutal reality—GitHub's real HTTP 406 errors for diffs over 3,000 lines (rolled out March 2024) were just the appetizer.
As devs, we're opinionated: AI slop isn't innovation; it's pollution. Tools like CodeRabbit are stepping up to auto-flag bugs, but why merge stochastic parrot vomit when forks are easier than ever? deckar01 on HN drops wisdom: "If you aren’t using your own code in production, you shouldn’t expect anyone else to." Amen. Anthropic's Claude spotting 500+ zero-days shows AI's power for good, but low-effort PR floods from projects like 8bitcn prove the flip side: burnout for volunteers.
Why This Matters for Your Workflow
- Detection is key: Scan for AI hallmarks—vague prose, no tests, zero docs. RFC 406i pushes lexical checks; pair it with GitHub's file-list APIs to bypass 406 limits.
- Reject smarter: Demand acceptance criteria for features, readable docs for fixes. Low bar? Sure, but human effort only.
- Tools to fight back: CodeRabbit reviews pre-merge; bots could auto-plonk slop. No more volunteer triage.
Critics whine it's "hostile" or CoC-violating (it parodies that too, reserving rights for "carbon-based entities capable of shame"). Bullshit. Genuine contributors bring value; bots bring noise. Over-rejection? Fork it yourself. In 2026's AI hype, this RFC is a wake-up: prioritize hygiene over quantity. OSS sustainability hangs in the balance—businesses delaying on large PRs post-406, maintainers quitting.
Bottom line: Adopt RFC 406i yesterday. It's not just funny; it's a blueprint for reclaiming our repos from the machines. Plonk the slop, devs—your sanity depends on it.
