15
Published Tools
15
Total Stars
0
Weekly Downloads
0
GitHub Followers
59
Public Repos
100/100
Avg Security
Published Tools
15 Skillsacross 5 categoriesKein-chronicle
Stop AI agents from reporting "absent" or "clean" on empty tool output: positive controls and a three-state result (present / absent / could not measure).
...moreKein-chronicle
Catch AI-content monoculture across a folder of posts and detect silent edits to an approved manuscript.
Kein-chronicle
Evaluate AI-generated art without vibes: position-bias-safe pairwise judging and similarity to reference images with open_clip.
...moreKein-chronicle
Safeguards for AI-assisted resume screening: lint rubrics for protected attributes, compute career facts in code, flag prompt injection in resumes.
...moreKein-chronicle
Audit a batch of AI-generated sprites: immutable style masters, generation ledger, duplicate frames, real transparency, plus safe background keying.
...moreKein-chronicle
Field-tested App Store Connect automation pitfalls: review-submission sequence, age-rating advertising flag, screenshot append bug, attachments lost on new versions.
...moreKein-chronicle
Detect grey-box screenshots: measure unique colors and flat-fill share of a rendered game canvas, and refuse to pass what it could not measure.
...moreKein-chronicle
Grade whether a video or page communicates, using independent cold readers checked against the author's truth file, and flag stale judgements.
...moreKein-chronicle
Lint and verify HTML motion scenes so any frame can be rendered exactly, in any order: no wall clock, no randomness, no timers.
...moreKein-chronicle
Machine QA for short videos with ffmpeg: frozen frames, dead air, loudness, true peak, size, a busy opening, and readable subtitle timing.
...moreKein-chronicle
Keep a license ledger for downloaded art and audio packs: fail on unknown packs, non-commercial licenses in a commercial release, and generate credits.
...moreKein-chronicle
Keep iOS simulator UI tests and crash-fuzz runs from dying on a laptop: the 16 GB memory wall and the crash-report dialog that never goes away.
...moreKein-chronicle
Let the LLM propose, let code judge: quote verification, number and date parsing, a rule evaluator, and a stable issue ledger for document review.
...moreKein-chronicle
Catch Swift runtime traps that unit tests cannot see: Int overflow, Int(Double.nan), bad casts and index errors, with a fuzz recipe and safe-arithmetic fixes.
...moreKein-chronicle
Catch web-game bugs the type checker can't: fonts that are declared but not shipped, runtime CDN dependencies, and page errors from headless Chromium.
...more