AI-first Game Core Engineer | Windows Maci. | OpenAI’s Ralphthon Winner ’26 | KOI Silver ’14 | Coding since age 9
lilmgenius builds agent skills that help an AI agent check its own work before and after acting: confirming it has understood a request (Readchk), stress-testing a plan with a single targeted objection (Hate), and auditing scattered facts to consolidate them into one source of truth (Ssotize). Their catalog also covers multi-lens review of artifacts (Prism), reordering lists and sections under a stated principle (Reorder), and making durable documents portable by generalizing tool-specific details (Detool). Their background as a game core engineer and a Ralphthon winner shows in skills aimed at engineering decision-making and reviewing plans.
Paperthin
28 skills · 1,375 stars · MIT
Aim When data lands with a thin ask — files, a repo, a paste, a chat scrollback — read it and propose…
Autobahn Carve guardrail-adjacent items out of scope with safe alternatives before risk-adjacent work star…
Catchup Rebuild the human's lost context on a project from live state, in plain language: what needs them…
Debloat Compress an artifact that has accreted into bloat — padding, over-qualification, fused sentences,…
Dedash Remove em-dashes and the dashes standing in for them from a user-owned scope, reading each occurr…
Detool Make durable artifacts portable by replacing incidental stack, vendor, model, CLI, path, quota, o…
Factchk Verify reality-grounded claims against external sources in both directions before they ship — cou…
Feynman Press a decision you just made until you can explain it to a skeptic in your own words, or surfac…
Hate Attack a plan, design, or argument like you want it to fail before you commit real effort — retur…
Macrothink User-invoked read-only pass for checking whether the current direction is tunnel-visioned: strip…
Mandela Audit any eval, metric, experiment, or benchmark for leakage — does external ground-truth enter i…
Modelchk Size a task's run before spending it: the cheapest sufficient capability tier (fast, standard, fr…
Nba Read the live cycle state and return the single highest-leverage next best action, not a menu.
Prism Split one artifact — a claim, plan, or file — across 2 to 5 independent lenses, one per genuinely…
Re0 Refresh an existing artifact into the current best v0.
Re0 Git Rewrite a finished commit's message into a clean, handoff-ready form in your own log style, so `g…
Re0 Loop Run repeated build -> QA -> re0-memo -> re0-work cycles while preserving learning and letting cod…
Re0 Memo Turn a finished, failed, or disappointing work cycle into portable lessons, anti-patterns, qualit…
Re0 Merge Review and land an external contribution the way this suite does: gate it against the thesis, lan…
Re0 Plan Open a paperthin iteration's casebook before re0-loop's first turn — seeded with real content the…
Re0 Release Walk a pending change through this repo's shipping and releasing checklist end to end, then tag a…
Re0 Upgrade Bring your installed paperthin skills up to the full current catalog in one step: retire what's b…
Re0 Work Restart a project or artifact from v0 while preserving only proven lessons, contracts, gates, voc…
Readchk Verify the model's understanding of a user's instruction before spending non-trivial work.
Reorder Realign a listing whose order has gone arbitrary — a list, table, catalog, sections, an enum — in…
Shower Cold-read the artifact you're focused on from a fresh, zero-context sub-session to confirm it sta…
Sip After you create or change an artifact or skill, taste-test it with our own skills instead of tru…
Ssotize Audit and consolidate a fact that's scattered across places into one canonical source, after appr…