← all publishers

zenithjuno

@zenithjuno source repo

Report this profile Claim this profile

• 23 skills
• 0 followers
• 2 hours ago last updated

1. ▌
   Handoff · zenithjuno bundle

   Create a Codex-first continuity handoff for continuing work in a fresh Codex session. Use whenever the user wants to continue in a new chat/session, preserve context, compact a long thread, avoid losing work, or leave a "start here" briefing. Triggers include "handoff," "hand this off," "continue in a new chat," "start a fresh chat," "this chat is getting long," "I'm running low on context," "summarize this so I can pick it up later," and Thai equivalents such as "ส่งต่อ," "สรุปไว้ไปคุยต่อแชทใหม่," "เปิดแชทใหม่," "แชทนี้ยาวแล้ว." Create repository-aware, scope-safe handoffs that preserve local storage conventions and current continuation routes while capturing conversation-only state, exact workspace paths, completed checks, settled decisions, and the immediate next action.
   0 installs
2. ▌
   Thai DOCX · zenithjuno bundle

   Create, edit, repair, audit, or preview Thai Word documents without mathematical notation: reports, letters, memos, school forms, minutes, and prose tables. Use TH Sarabun New with editable, insertion-safe Thai text; repair legacy Thai fonts in imported DOCX. For mathematical notation or editable equations, use thai-math-docx instead. Administrative numbers and prose such as คะแนน ≥ 80 belong here.
   0 installs
3. ▌
   Playwright · zenithjuno bundle

   Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.
   0 installs
4. ▌
   Ship Check · zenithjuno bundle

   Use as the final read-only gate before publishing or sharing an educational ("สื่อการเรียนรู้") HTML/JS browser game made by a solo non-developer. It answers one question: "is this safe to put in front of learners yet?" Three parts: (1) a static scan script (scripts/ship_static_scan.py, stdlib only) that flags ship-blockers in the HTML/JS — dev leftovers, insecure/localhost URLs, missing mobile/charset/lang meta, external network dependencies that break offline use; (2) a functional dry-run — the AI plays the game live if a browser/preview tool is available, otherwise a static review plus a manual dry-run checklist the user runs; (3) a readiness checklist that confirms the other skills are green (content audited, playtested, objective deliverable, player data works). Produces a PASS / CONCERNS / FAIL verdict and never edits game files.
   0 installs
5. ▌
   Whats Next · zenithjuno bundle

   Use when the user is unsure what to do next on their educational ("สื่อการเรียนรู้") HTML/JS browser game, or is starting out and wants orientation. This is the map for the learning-game toolset. It is READ-ONLY: it inspects the project (design spec, game files, playtests, reports), figures out which lifecycle stage the project is in, and recommends ONE concrete next action and the skill to run for it. Designed for a solo non-developer — plain language, no jargon, never more than one main recommendation. Use on phrases like "what now", "where am I", "ทำอะไรต่อดี", "เริ่มยังไง", "next step".
   0 installs
6. ▌
   Task Scoping · zenithjuno bundle

   Use before non-trivial coding work to turn the request into a compact task contract and choose the lightest safe workflow: S for a clear local change, M for a short direction-and-verification plan, or L for a substantial, high-risk, ambiguous, multi-system, or multi-session build routed to grill-to-build. Trigger on coding features, fixes, refactors, migrations, or requests that need direction before editing but may not justify a full design process. Skip genuinely obvious one-line changes.
   0 installs
7. ▌
   Skill Release · zenithjuno bundle

   Publish any update, addition, repair, rename, or removal of a user-managed Codex/Claude skill from the local `~/.codex/skills` source checkout to its GitHub mirror. Use whenever an agent changes a skill folder or asks to commit, push, install, sync, release, or make a skill available to other sessions. Enforce scoped preflight, verification, commit/push, and clean local-mirror proof through the shared release helper.
   0 installs
8. ▌
   Grill To Build · zenithjuno bundle

   A rigorous, coding-first design-before-build process for Mode L substantial, high-risk, ambiguous, multi-system, or multi-session work. The agent grills the user with recommendation-backed questions, locks decisions in a live ledger, produces a BLUEPRINT (what) and CONSTRUCTION_PLAN (how and tests), and refuses to build before explicit approval. For coding it establishes bounded-context BUILD-CONTROL, exact path scopes, current contract indexes, AGENTS.md routing, checkpoints, and cold audit logs. Use when task-scoping routes work to L, when costly product decisions remain, or when the user explicitly asks "grill me", "plan this first", "spec this", "grill-to-build", or "deep-grill-to-build". Do not impose the full artifact set on clear one-session S/M coding tasks.
   0 installs
9. ▌
   Thai Math DOCX · zenithjuno bundle

   Use when creating, editing, repairing, auditing, or reconstructing Thai mathematics Microsoft Word .docx files. Optimize for the user's Word workflow: Thai prose in TH Sarabun New 16 pt Complex Script, Latin/admin prose in Cambria 12 pt, editable Word Equation/OMML for math-ish tokens, insertion-safe Thai body runs, Thai-style question labels, TH Sarabun New 12 pt footers, robust docDefaults/Normal style defaults that survive Clear Formatting, and mandatory font/OMML audits. Use for Thai math exams, math handouts and answer keys, PDF/image-to-DOCX reconstruction of mathematical documents, and repair of imported Thai DOCX that contain equations — any task where Thai text plus mathematical notation must remain editable and Word-compatible. For a Thai .docx with NO mathematical notation (plain prose, letters, memos, reports, forms, plain handouts, or repairing an imported document that has no equations), use the `thai-docx` skill instead.
   0 installs
10. ▌
    Build Changelog · zenithjuno bundle

    Use during the BUILD phase of a Mode L planned coding project to operate bounded-context build control: keep one hot `BUILD-CONTROL-{slug}.md` containing the Project Map, current stage, version-control coordinates, active-contract routing, open changes, and cold-log index; append PRG/CHG audit entries to phase-sharded `BUILD-LOG` files; enforce addressed stage/change gates; and checkpoint only declared repository paths. Prefer it whenever coding from a BLUEPRINT/CONSTRUCTION_PLAN, resuming a long build, avoiding repeated full-log reads, recording staged progress, or handling a deviation. Apply the same current-state-versus-cold-history principle more lightly to non-coding staged work. Triggers include "start/resume the build", "build log", "build control", "record this stage/change", "บันทึก build", and "continue the construction plan".
    0 installs
11. ▌
    Plan Scrutinize · zenithjuno bundle

    Outsider cold-read of a plan, BLUEPRINT, or design spec BEFORE any building, to catch the expensive errors — building the wrong thing, scope that drifted past the original problem, assumptions that break against reality — while a fix still costs one sentence. Pairs with grill-to-build: grill-to-build authors the plan collaboratively; plan-scrutinize reads the finished plan cold and pressure-tests it as an outsider. Trigger on /plan-scrutinize and proactively whenever the user wants to sanity-check, pressure-test, cold-review, audit, or get a second opinion on a plan, blueprint, spec, or design BEFORE executing it — including Thai phrasings like "ตรวจแผนก่อน", "scrutinize แผน", "เช็คแผนก่อนลงมือ", "sanity check แผนหน่อย", "แผนนี้มีอะไร drift ไหม", "พร้อม build ยัง". Do NOT use this to author a plan (that is grill-to-build) or to review an already-built artifact (that is a post-build review). This reviews the plan, not the product. Always writes a standalone PLAN-SCRUTINY-<slug>.md report — the shared medium th
    0 installs
12. ▌
    Learner Playtest · zenithjuno bundle

    Use to collect and analyze HUMAN feedback on an educational ("สื่อการเรียนรู้") browser game, to guide further development. Three modes: 'survey' produces a ready-to-paste Google Form question set (learner self-report + a short comprehension check) that the user pastes into Google Forms by hand — no API, no dependency; 'observe' produces a form for a teacher to fill in while watching one learner play; 'analyze' turns Google Form responses (exported to a Sheet/CSV) or session notes into a structured report with a verdict. The central question is whether THIS learner actually learned the target thing and where they got stuck — not just whether it was fun. The skill itself does not simulate a player and does not create the live Google Form; it writes question sets and analyses, and writes report files only on approval.
    0 installs
13. ▌
    Project Bootstrap · zenithjuno bundle

    Establish or adopt project entrypoints and context routes, or audit project context health and unnecessary reading. Use for new-project setup, project onboarding, broken resume paths, conflicting current sources, or an explicit context/token-efficiency audit. Coding-first with a generic routing core. A supplied snippet or routine edit needs no bootstrap or full audit.
    0 installs
14. ▌
    Game Content Audit · zenithjuno bundle

    Use to audit the content and data of an educational ("สื่อการเรียนรู้") browser game BEFORE shipping or after changing the engine. Two targets: (1) the HTML/JS question-generator code — reviewed by reading it for coverage, difficulty spread, and especially answer-correctness; and (2) data files — a sample of generated questions and the player-stat table (e.g. exported from Google Sheet) — checked mechanically by the bundled Python script (scripts/audit_game_content.py, stdlib only, no node). This skill is READ-ONLY: it produces a PASS / CONCERNS / FAIL report and never edits game files. Anything it cannot verify it flags as MANUAL CHECK NEEDED rather than assuming pass.
    0 installs
15. ▌
    Soffice Runtime Fix · zenithjuno bundle

    Diagnose and repair the bundled macOS LibreOffice runtime when DOCX-to-PDF conversion, dylib loading, writable profiles, or Thai font rendering fails after a Codex runtime update. Classify DOCX-to-PDF separately from PDF-to-PNG failures; do not patch LibreOffice for a missing rasterizer alone.
    0 installs
16. ▌
    Thai Font Normalize · zenithjuno bundle

    Normalize Thai fonts in generated Word .docx files to TH Sarabun New so that Thai text renders correctly in Word, in PDF export, and after Clear Formatting or applying paragraph styles. Use this skill WHENEVER generating, editing, or producing a .docx that contains Thai text — even if the user does not explicitly mention fonts. This prevents the common failure where Thai falls back to Angsana New. The skill both (a) writes correct OOXML during generation and (b) runs a bundled verification script on the finished file as a mandatory final pass. Default Thai font: TH Sarabun New, Complex Script size 16pt.
    0 installs
17. ▌
    Feedback To Leverage · zenithjuno bundle

    Use when a human corrects, rejects, redirects, or repeatedly fixes coding work, documentation, architecture, workflow, or tool usage. Fix the immediate issue, classify why it happened, and place the smallest justified durable safeguard in a test, static rule, source-of-truth document, project map/index, decision record, template, or helper script. Integrate with build control without creating a new feedback log. Skip system changes for genuinely one-off preference corrections.
    0 installs
18. ▌
    Learning Game Design · zenithjuno bundle

    Use when designing an educational ("สื่อการเรียนรู้") game through a discussion-first, lightweight design workflow. The user is a non-developer teacher/author building a browser game (HTML/JS front-end, optional Google Sheet for player stats). This skill helps think through the design in chat — learning objective, target learner, core loop, content architecture, minimal tech, and how success will be judged — then writes ONE lean design spec only after explicit approval. It is deliberately lightweight: not a full Game Design Document, no studio pipeline. Bias every decision toward minimal dependencies, low spec, and something a solo non-dev can maintain.
    0 installs
19. ▌
    Math Handout Sandbox · zenithjuno bundle

    Design and review Thai mathematics teaching materials through a teacher-led, discussion-first workflow. Use when a teacher brings images, worksheets, worked or faded examples, exercise lists, handout drafts, or lesson ideas and wants a quick pedagogical review, a project-aware design discussion, or approved production routing. Analyze teaching flow, misconceptions, mathematical correctness, and variants before creating an SVG, DOCX, slide, or worksheet. Do not use for direct Thai DOCX repair or one-off formatting.
    0 installs
20. ▌
    Systematic Debugging · zenithjuno bundle

    Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes
    0 installs
21. ▌
    Blind Answer Key Audit · zenithjuno bundle

    Audit an answer key (exam solutions, model answers, a generated question bank) for correctness by BLIND re-solving — an independent solver answers each item WITHOUT seeing the provided key, then diffs against it, marking agreements as pass and disagreements as flags for a human to adjudicate. Use this WHENEVER you need to verify or QA answers/solutions produced by someone else (another AI such as Codex, a textbook, a teacher, an auto-generator) and correctness matters — to catch wrong keys, contradictory questions, no-valid-option items, and duplicate/typo choices before learners or users see them. Triggers include: "ตรวจเฉลย", "audit answer key", "verify these solutions", "blind check", "cross-check the answers", "QA this question bank", "is this answer key correct", and any producer↔checker AI workflow. Roles are swappable — the same process works whether THIS agent produced the answers or is the one checking them. Prefer this over ad-hoc "just re-grade it" because the blind discipline is what prevents anch
    0 installs
22. ▌
    Thai Math Exam Production · zenithjuno bundle

    Design and manage Thai mathematics exams and parallel sets from approved exams: teacher-approved format, chapter-specific difficulty, blueprint, item variants, working solutions, and whole-paper review. Keep structured exam state and a teacher-readable EXAM-DESIGN.md; route approved content to blind audit and Thai DOCX production. Not for ordinary worksheets, direct formatting, or standalone answer-key checking.
    0 installs
23. ▌
    Verification Before Completion · zenithjuno bundle

    Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
    0 installs