Skip to main content
tmdgusya
GitHub creator profile

tmdgusya

Repository-level view of 65 collected skills across 25 GitHub repositories.

skills collected
65
repositories
25
updated
Jul 26, 2026
Showing the top 8 repositories here; full repository list continues below.
repository explorer

Repositories and representative skills

clarification
software-developers

Use when a user's request is vague, ambiguous, or underspecified. Explores the codebase first, then runs an iterative Q&A loop grounded in the findings until ambiguity is gone. Outputs a clear, well-scoped context brief so the user can plan sharply. Triggersโ€ฆ

Jul 2, 2026
clean-ai-slop
software-developers

Corrective cleanup of AI-generated code โ€” removes LLM-specific patterns while preserving behavior. Use when the user says "deslop", "slop", "clean AI code", "remove AI patterns", or when you spot LLM-generated code smells after any generation session. Forโ€ฆ

Jul 2, 2026
long-run
software-developers

Orchestrates multi-day execution of complex tasks through milestones. Each milestone goes through plan-crafting, run-plan (worker-validator), and review-work phases with checkpoint/recovery. Runs autonomously end-to-end โ€” no per-milestone approval gates;โ€ฆ

Jul 2, 2026
milestone-planning
software-developers

Decomposes complex, multi-day tasks into optimized milestones (ultraplan). Broad parallel exploration grounds a draft milestone DAG, then independent adversarial critics attack the draft in rounds until no blocking findings remain. Triggers when the user saysโ€ฆ

Jul 2, 2026
plan-crafting
software-developers

Use when a task's scope is clear and multi-step implementation is needed, before touching code. Triggered after clarification is complete, or when the user explicitly requests plan creation with a clear prompt.

Jul 2, 2026
review-work
software-quality-assurance-analysts-and-testers

Use after run-plan completes to independently verify the implementation. Reads only the plan document and inspects the codebase from scratch โ€” information-isolated from the execution context. Produces a structured review document with PASS/FAIL verdict.โ€ฆ

Jul 2, 2026
run-plan
software-developers

Use when you have a written implementation plan to execute. Loads the plan, reviews critically, executes tasks in dependency order, and reports completion. Triggers when the user says "run the plan", "execute the plan", or "let's start implementing".

Jul 2, 2026
systematic-debugging
software-developers

Use when encountering any bug, test failure, or unexpected behavior. Enforces a strict reproduce-first, root-cause-first, failing-test-first debugging workflow before fixing.

Apr 5, 2026
Showing 8 of 11 collected skills.
agentic-clarification
computer-occupations-all-other

Use when a user's request is vague, ambiguous, or underspecified. Runs recon-first clarification โ€” codebase exploration before questions, then one bundled question round with defensible defaults โ€” and outputs a Goal Contract for the durable /goal runtime.โ€ฆ

Jul 8, 2026
agentic-goal
computer-occupations-all-other

Primary execution workflow for durable /goal runtime. Use when a Goal Contract is active or when the user asks to execute, continue, verify, or complete a goal.

Jul 8, 2026
agentic-simplify
software-developers

Review changed code for reuse opportunities, quality issues, and inefficiencies using three parallel review agents, then fix any issues found. Triggers when the user says "agentic-simplify", "clean up the code", "review the changes", or after goalโ€ฆ

May 28, 2026
agentic-systematic-debugging
software-quality-assurance-analysts-and-testers

Use when encountering any bug, test failure, or unexpected behavior. Enforces a strict reproduce-first, root-cause-first, failing-test-first debugging workflow before fixing.

May 28, 2026
agentic-brainstorming
project-management-specialists

Interactive idea development through guided Q&A dialogue. This skill helps users clarify and develop their ideas by asking targeted questions, expanding on possibilities, and producing a structured markdown document capturing the essence of their thinking.โ€ฆ

May 8, 2026
agentic-karpathy
software-developers

Behavioral guardrails to prevent common LLM coding mistakes โ€” enforces surgical changes, assumption verification, and scope discipline before and during implementation. Use when implementing features, modifying code, or when you notice yourself about to makeโ€ฆ

May 5, 2026
agentic-rob-pike
software-developers

Rob Pike's 5 Rules of Programming โ€” a decision framework that prevents premature optimization and enforces measurement-driven development. Use when the user says "optimize", "slow", "performance", "bottleneck", "speed up", "make faster", "too slow", or anyโ€ฆ

Apr 5, 2026
blueprint
software-developers

Light planning mode. Use when the user requests a small, clearly-scoped change (roughly โ‰ค3 files, no new interfaces, no ambiguity) that still deserves a written plan before code. For ambiguous or multi-component work, use the forge skill instead. Produces anโ€ฆ

Jul 12, 2026
forge
software-developers

Strong planning mode. Use when the user requests a non-trivial feature, refactor, or system change โ€” especially when the request is ambiguous, spans multiple files, or says "์ œ๋Œ€๋กœ", "๊ฐ•ํ•˜๊ฒŒ", "๋นก์„ธ๊ฒŒ", "thoroughly", "strong plan". Runs deep codebase recon BEFOREโ€ฆ

Jul 12, 2026
hammer
software-developers

Implementation loop. Use when a plan document exists (from forge or blueprint) and implementation should run โ€” including when the user says "implement", "execute the plan", "๊ตฌํ˜„ํ•ด", "์‹คํ–‰ํ•ด", or approves a plan. Executes tasks through a worker โ†’ validator โ†’โ€ฆ

Jul 12, 2026
pptxx
graphic-designers

Presentation deck mode. Use when the user wants a presentation, slide deck, or PPTX built from a topic or script โ€” especially when they say "๋ฐœํ‘œ", "์Šฌ๋ผ์ด๋“œ", "ํ”„๋ ˆ์  ํ…Œ์ด์…˜", "๋ฐœํ‘œ ์ž๋ฃŒ ๋งŒ๋“ค์–ด์ค˜", "์Šฌ๋ผ์ด๋“œ ๋ฝ‘์•„์ค˜", "presentation", "slide deck", "pptx", "deck". Secures a script,โ€ฆ

Jul 7, 2026
crucible
graphic-designers

Storyline-driven design mode. Use when the user makes a storyline- or reference-driven design request โ€” design tokens, mood/visual identity, a style guide โ€” especially when they say "๋””์ž์ธ", "๋ฌด๋“œ", "์Šคํ† ๋ฆฌ", "๋ ˆํผ๋Ÿฐ์Šค", "design tokens", "style guide". Prospects realโ€ฆ

Jul 7, 2026
align
software-developers

๊ตฌํ˜„ ์ „์— ์˜๋„ ์ •๋ ฌ์„ ์ ˆ์ฐจ๋กœ ๊ฐ•์ œํ•œ๋‹ค. ์ฝ”๋“œ๋ฒ ์ด์Šค๋ฅผ ๋จผ์ € ํƒ์ƒ‰ํ•˜๊ณ  ๋ชจ๋“  ๋ชจํ˜ธํ•จ์„ ๋„์ง‘์–ด๋‚ด ์‚ฌ์šฉ์ž์™€ ํ•ฉ์˜ํ•  ๋•Œ๊นŒ์ง€ ๋ฐ˜๋ณตํ•œ ๋’ค, ์ •๋ ฌ๋œ ์˜๋„๋ฅผ handoff ๋ฌธ์„œ(์˜๋„ + ํ™•์ • ๋ฒ”์œ„ + ๊ฒ€์ฆ ๊ฐ€๋Šฅํ•œ ์™„๋ฃŒ ์กฐ๊ฑด)๋กœ ๋‚จ๊ธด๋‹ค. ์‚ฐ์ถœ๋ฌผ์€ ์ฝ”๋“œ๊ฐ€ ์•„๋‹ˆ๋‹ค. ๋‹ค๋‹จ๊ณ„ ๊ตฌํ˜„ ์ „, ์š”์ฒญ์ด ๋ชจํ˜ธํ•  ๋•Œ, ๋˜๋Š” ์‚ฌ์šฉ์ž๊ฐ€ "align", "์ •๋ ฌ", "์˜๋„ ๋งž์ถฐ", "๋ฌธ๋งฅ ํŒŒ์•…ํ•ด", "๋ช…ํ™•ํžˆ ํ•ด์ค˜"๋ผ๊ณ  ํ•  ๋•Œ ์‚ฌ์šฉํ•œ๋‹ค. ์Šค์Šค๋กœ ํŠธ๋ฆฌ๊ฑฐ๋กœ ์“ธ ์ˆ˜๋„ ์žˆ๋‹ค โ€” ์š”์ฒญ์ด ๋ชจํ˜ธํ•ด ๋ณด์ด๋ฉด ์ฐ์–ด์„œโ€ฆ

Jun 17, 2026
forge
computer-occupations-all-other

align์ด ๋งŒ๋“  handoff ๋ฌธ์„œ์˜ "์™„๋ฃŒ" ์กฐ๊ฑด์„ ZCode /goal์ด ํŒ์ •ํ•  ์ˆ˜ ์žˆ๋Š” ๊ฐ๊ด€์  ์ธ์ˆ˜์กฐ๊ฑด(์‹คํ–‰ ๋ช…๋ น + ์˜ˆ์ƒ ๊ฒฐ๊ณผ ์Œ)์œผ๋กœ ๋ณ€ํ™˜ํ•ฉ๋‹ˆ๋‹ค. ์‚ฐ์ถœ๋ฌผ์€ ์ธ์ˆ˜์กฐ๊ฑด ๋ชฉ๋ก์ด๋ฉฐ ์ฝ”๋“œ๋Š” ์“ฐ์ง€ ์•Š์Šต๋‹ˆ๋‹ค. handoff๊ฐ€ ์žˆ์„ ๋•Œ, ๋˜๋Š” ์‚ฌ์šฉ์ž๊ฐ€ "forge", "์ธ์ˆ˜์กฐ๊ฑด ๋งŒ๋“ค์–ด", "๊ฒ€์ฆ ์กฐ๊ฑด ์„ธ์›Œ"๋ผ๊ณ  ํ•  ๋•Œ ์‚ฌ์šฉํ•˜์„ธ์š”.

Jun 17, 2026
prometheus
computer-occupations-all-other

๊ฒ€์ฆยท์™„์ˆ˜ยท์กฐ์‚ฌ๋ฅผ ์ ˆ์ฐจ๋กœ ๊ฐ•์ œํ•ด์„œ GLM-5.2(ZCode์˜ ๋ชจ๋ธ)๊ฐ€ ์ž‘์—…์„ ๋๊นŒ์ง€ ๋ฐ€๊ณ  ๊ฐ€๊ฒŒ ๋งŒ๋“œ๋Š” ํ•˜๋„ค์Šค. ๊ฐ•ํ•œ ์—์ด์ „ํŠธ ๋ชจ๋ธ์˜ ์ž‘์—… ์Šต๊ด€์„ GLM-5.2๋กœ ๊ฐ€์ ธ์˜จ๋‹ค. ๋‹ค๋‹จ๊ณ„ ์ž‘์—…(2๊ฐœ ์ด์ƒ์˜ ์ˆœ์ฐจ ๋ชฉํ‘œ), ๊ธด ์ž์œจ ์ž‘์—…, ๋””๋ฒ„๊น…/์›์ธ ์กฐ์‚ฌ, ๋ Œ๋”ยท์‹คํ–‰ ์‚ฐ์ถœ๋ฌผ(HTML, SVG, ๊ฒŒ์ž„, ์ฐจํŠธ)์„ ๋งŒ๋“ค ๋•Œ, ๋˜๋Š” ์‚ฌ์šฉ์ž๊ฐ€ "prometheus", "๋๊นŒ์ง€ ํ•ด์ค˜", "verify as you go", "๋ชฉํ‘œ๋กœ ์ชผ๊ฐœ์ค˜"๋ผ๊ณ  ํ•  ๋•Œ ์‚ฌ์šฉํ•œ๋‹ค. ์ฆ๊ฑฐ ์—†๋Š” "์™„๋ฃŒ"๋ฅผโ€ฆ

Jun 17, 2026
Showing 12 of 25 repositories