Loading...
Loading...
Found 3,053 Skills
Validates dataset formatting and quality for SageMaker model fine-tuning (SFT, DPO, or RLVR). Use when the user says "is my dataset okay", "evaluate my data", "check my training data", "I have my own data", or before starting any fine-tuning job. Detects file format, checks schema compliance against the selected model and technique, and reports whether the data is ready for training or evaluation.
Personalized links without explicit login for authentication. Apply when implementing authentication flows that don't require traditional login.
Run tests against a validated RootSpec specification and report results. Use this when a user wants to run tests, check what's passing, get a test report, or verify their implementation works.
Challenges AI-generated plans, code, designs, and decisions before you commit. Pairs with any other skill as a review layer. Uses pre-mortem analysis, inversion thinking, and Socratic questioning to find what AI missed — blind spots, hidden assumptions, failure modes, and optimistic shortcuts. The skill that asks "are you sure about that?" so you don't have to. Triggers on: "challenge this", "devils advocate", "stress test this plan", "what could go wrong", "poke holes in this", "review this critically", "second opinion on this design", "what am I missing". Use this skill when you need critical review of any AI-generated output, architecture decision, implementation plan, or code before committing to it.
Accepts Triton operator implementations, automatically invokes Torch small operator implementations (CPU or NPU) for precision comparison, and generates precision reports. It is used when users need to verify the correctness and precision of Triton operator implementations, compare precision with PyTorch implementations, and generate standardized precision reports.
Process and transform markdown documents. Convert to HTML, extract headings, generate TOC, check links. Use when working with documentation or markdown content.
기존 PeachSolution 모듈을 test-data 패턴으로 리팩토링하는 팀 오케스트레이터 스킬. "팀 리팩토링", "레거시 코드 정리", "test-data 패턴으로 변환" 키워드로 트리거. layer=backend|frontend|all 지원, 독립 QA로 확증 편향 방지.
Adversarial stress-test of a /think intelligence brief. Reads the think output markdown, then deploys 5-7 of the same analytical frameworks — but each one is hunting exclusively for reasons the recommendation is wrong, the conviction is unearned, and the idea will fail. Every framework becomes a prosecutor, not a judge. Surfaces the strongest kill shots, identifies which parts of the original brief are load-bearing but unverified, and produces a Red Team Report with a survival verdict. Use when the user says "red-team this", "attack this", "poke holes", "steel-man the opposition", "why is this a bad idea", "/red-team", or presents a /think brief they want stress-tested.
Neverbounce integration. Manage data, records, and automate workflows. Use when the user wants to interact with Neverbounce data.
Use when OpenSpec artifacts have been generated by /opsx:propose and need review before implementation begins — validates proposal scope, spec completeness, design decisions, and task executability
Review storyboard prompts and video-request prompts before generation. Use this when a prompt draft already exists and you need to catch weak first frames, drift risk, missing constraints, bad product timing, or generic ad-like language before spending model credits.
Migration Architect