Showing 6961–6990 of 10165 skills
AutoGluon preset selection strategy: when to use medium/good/high/best_quality, when EDA is unnecessary, and how to tune within a preset. Validated on s6e7 (AG high_quality 600s → OOF=0.8739, LB=0.87458, gap=0.0007 — perfect alignment). Use when: (1) Deciding which AutoGluon preset to start with, (2) Wondering whether to do manual EDA before fitting, (3) Wanting to know which hyperparameters to tune first, (4) Deciding time budget for tabular training. Differs from `autogluon-first` (which says "just use best_quality"): this skill explains the preset tradeoffs and where manual intervention pays off vs not.
topprismdata/cultivating-ml-agent
Extract reusable knowledge from work sessions into new Claude Code skills. Use when: (1) /claudeception command to review session, (2) "save this as a skill" or "extract a skill", (3) "what did we learn?", (4) After non-obvious debugging, workarounds, trial-and-error discovery, or counterintuitive solutions. Do NOT trigger for routine fixes or straightforward tasks with obvious solutions.
topprismdata/cultivating-ml-agent
Use when context window is getting crowded, when dealing with long training logs, or when preparing prompts with mixed-priority content (system / task / skills / examples). Triggers when assembling prompts >4000 chars, or when observing "Lost in the Middle" symptoms (model ignores mid-prompt content).
topprismdata/cultivating-ml-agent
Use when: (1) You have limited daily submission quota and 50+ skill files to explore, (2) Most Kaggle experiments cost 1 quota each but only the 1-2 BEST actually matter for your final score, (3) You waste hours doing redundant experiments because the EXISTING public kernels (rank 4-9 players) have already solved 80% of the problem, (4) You're tempted to re-submit "to see what happens". Key insight from 4+ months, 20+ competitions: **0.6 × BEST_PUBLIC ≥ 0.8 × OPTIMAL_OWN**, i.e. the best public kernel + small tuning > optimal own work in 90% of cases. Most Kaggle competition success comes from picking the right public kernel and spending quota wisely, not from complex custom work.
topprismdata/cultivating-ml-agent
Use when: (1) You've completed a competition or experiment and want to extract reusable knowledge, (2) Your agent keeps repeating the same mistakes across competitions because lessons weren't crystallized, (3) You have 50+ memory files but can't find relevant knowledge when starting a new task, (4) You need to decide what to remember vs what to forget. Covers the complete crystallization cycle: Experiment → Identify Outcome → Extract Pattern → Classify (feedback/ learned/project/reference) → Store with triggers → Activate on match. Includes the "3-layer architecture" (L1 core, L2 domain skills, L3 cross-domain principles) and the "forgetting protocol" for stale knowledge. Validated across 4+ months, 20+ competitions, 109+ crystallized memories.
topprismdata/cultivating-ml-agent
Use when extracting entities and relationships from text, when building dynamic knowledge graphs, when wanting to upgrade from static OKF to entity-aware memory, or when analyzing papers/discussions for concepts. Triggers when ingesting new papers, processing Kaggle writeups, or when existing skills can't capture connections between concepts.
topprismdata/cultivating-ml-agent
Use when context window is getting crowded, when agent needs to recall past experiences, or when starting a long-running task that spans multiple sessions. Triggers when you have 43+ skills and can't fit them all in context, when you need to decide "what should be in working memory right now", or when persisting new learnings for future sessions.
topprismdata/cultivating-ml-agent
Augment a BC Transformer agent with 2-ply bounded search using the cg library's SearchBegin/SearchStep API. Validated 2026-07-30 on Pokemon TCG AI Battle: search-enhanced agent submitted as v8 (LB pending). Use when: (1) you have a trained BC policy/value model, (2) the cg library exposes a search API (SearchBegin/SearchStep/SearchRelease), (3) you want lookahead beyond pure reaction-based action selection. Key risks: hidden-info sampling for unknown opponent state (use SNORLAX=1072 pad by convention), time budget per decision (N=5 samples × ~30 search_steps ≈ 1-3s), value-head quality unknown without calibration. Start with small N_SAMPLES and short rollout, tune up.
topprismdata/cultivating-ml-agent
Submit a BC (behavioral cloning) Transformer agent for the Pokémon TCG AI Battle competition. Validated 2026-07-30: best.pth (15-epoch BC Large, d_model=256) → **public LB 600.0** (vs prior 165.6 — 3.6× improvement from using the best checkpoint instead of a stale submission). Use when: (1) training a BC agent on Kaggle Pokémon TCG episodes, (2) submitting a tar.gz agent for the competition, (3) scaling BC data to 80K+ games. Critical pitfall: your highest validation WR checkpoint is almost certainly NOT what you last saved to your submission directory — always verify the submission's model_bc.pth matches the best checkpoint before submitting. Stream-sampling (load 50K per part then drop) prevents OOM when scaling to 500K+ samples on a 64GB machine.
topprismdata/cultivating-ml-agent
Use when running Python code via sandbox, when an experiment fails with a runtime error, when building self-correcting ML agents, or when debugging auto-generated code. Triggers on subprocess failures, ImportError, CUDA OOM, shape mismatch, or any "agent gets stuck on syntax error" pattern.
topprismdata/cultivating-ml-agent
Builds the person's three.js request over a set time (an hour, two hours, an afternoon) from a short brief in paprika-loop/PROMPT.md, which they can change while it runs, in a loop of fresh sessions, each as many improvements as fit, judged against the brief, scored and committed until the time is up. Use on /paprika-loop <request> for <time>, "keep improving this for two hours", or any 3D build the person wants done unattended over a long time.
Keeno-Corp/paprika3d
Profile a three.js or WebGL page from a coding agent - frame time (fps), draw calls, triangles and asset size against a budget, and which objects or settings cost the frame. Use when it stutters, has low fps or loads slowly. paprika3d's perf tool.
Keeno-Corp/paprika3d
Play a three.js or WebGL game or interactive page from a coding agent - keys, mouse, touch and scroll on exact frames of a paused, seeded clock, the state read back as JSON. Use to playtest its controls. paprika3d's interact tool.
Keeno-Corp/paprika3d
Compact the current conversation into a handoff document for another agent to pick up.
leek/agent-skills
Recover when the AGENT itself is stuck — looping, repeating the same failed action, burning tokens/budget, overflowing context, drifting from the goal, or acting on stale state. Use when a run isn't converging and reworded retries aren't helping. This is a workflow you follow, not a hidden runtime that auto-heals. For bugs in the product code use fixing-bugs.
Cristhianzl/claude-skills-czl
Check whether an App Store submission meets Apple's stated criteria for an expedited review, draft the request text (critical bug or time-bound event), and give the link to Apple's request form. Use when the user says "expedite my review", "speed up App Store review", "urgent fix needs to go live", "my launch event is in a few days and the app is still in review", or asks whether Google Play reviews can be expedited. This skill prepares the request. It cannot file it and cannot make a review faster.
deonmenezes/mobile-app-skill
阿杭AI博主系列视频制作入口。用于AI博主口播剪辑、剪后逐字稿和对应封面,选择本系列参数后调用另行安装的剪辑引擎与封面Skill。
ahang008/ah-skills
Run fast, progressive experiments to map tunable parameters and their effects. Use when optimizing code, prompts, configurations, or other artifacts against a goal or benchmark, or when a long research run needs a planned hypothesis queue, focused trials, and durable findings.
transitive-bullshit/skills
Discover, compare, install or update agent skills when the user explicitly seeks skills or asks to extend their skill collection. Use for skill-discovery requests, not ordinary how-to or implementation questions.
transitive-bullshit/skills
a5c-ai/babysitter
a5c-ai/babysitter
AI 电影/短剧解说视频自动生成(AI 解说大师 CLI Skill)。当用户需要创建电影解说视频、短剧解说、影视二创、AI 配音旁白视频、film commentary、video narration、drama dubbing、movie narration 时触发。内置电影素材库、BGM、多语种配音、解说模板。通过 narrator-ai-cli 命令行实现:搜片→选模板→选 BGM→选配音→生成文案→合成视频的全流程自动化。CLI client for Narrator AI video narration API.
NarratorAI-Studio/narrator-ai-cli-skill
Smart router for stack skills, or senior fullstack for ad-hoc scripts/HTML/automation. Use for generic development or when invoking /developer.
tibursocampos/agent-dev-toolkit
Present the installed static skills catalog (CATALOG.md + OPERATOR.md) without loading every SKILL.md. Use when listing skills or invoking help-skills (host: /help-skills, $help-skills, use skill help-skills, or OpenCode skill tool).
tibursocampos/agent-dev-toolkit
Create a GitHub pull request with gh CLI against the repository's base branch, template resolution, mandatory title/body confirmation, and mandatory auto-merge ask. Use whenever the user wants a PR opened — including /open-github-pr, open PR, create pull request, abrir PR, criar PR, abrir o PR, faça o PR, fazer o PR, open the PR, open a pull request, fluxo completo (when PR is in scope), commit+push+PR, push and open PR, release PR, or any handoff from /commit or /push that implies opening a pull request. Never create a PR with ad-hoc gh commands outside this skill; follow this SKILL.md end-to-end.
tibursocampos/agent-dev-toolkit
Use when 用户要用 Gemini 原生多模态联合分析本地或在线视频的画面、声音与时序,支持 YouTube 直读,以及按需下载 B站、小红书、抖音或经一次性授权解析微信视频号;触发词:分析视频、录屏质检、视频链接分析、批量验收。只要字幕、只下载、或要剪辑出片(交给 sandy-video)时不用此 Skill。
sanshengai/sansheng-gemini-video
Write a contract-style prompt for a hard, precisely decidable problem: an open conjecture, a prove-or-disprove task, or a beat-the-frontier optimization. Use when ordinary prompting returns hedged partial results, special cases, surveys, "this reduces to another open problem", or early give-ups on long autonomous runs.
odinfree/storm-ecdsa-harness
Akashneelesh's method on ecdsa.fail (17 rows, Jun-Nov 2026; Q765 from 128.7M to 48.4M Toffoli in 37 hours): take a cheaper public walk a few qubits wider and buy the qubits back with a page of low bits whose fourth word is implicit in the walk's invariant, narrow end machines shipped as checked gate lists, a staged build where switch-off rebuilds the parent, and a census failure budget. Use when a public walk k qubits wider costs a fraction of your line; when a two-ring Euclid walk must lose wires; when a stream is assembled from modules built by separate agents; when summing a failure budget term by term; or when you review a row built this way.
odinfree/storm-ecdsa-harness
郑希观点库——基于易方达基金经理郑希 2012–2026 年全部公开观点原文语料,外加从语料蒸馏、有本人原话佐证的郑希投资方法的可溯源 research skill。能做: (1) 溯源问答——他怎么看 AI算力/光通信/新能源/半导体/ROE 等,引用其原话作答; (2) 讲解他的投资方法/框架/选股逻辑;(3) 前瞻应用——用他的方法分析当下任意主题/行业/个股,语料没谈过也能据框架推演; (4) 风格化点评——用他季报/手记的口吻写市场点评、季度展望; (5) 言行对照——用他全部基金真实数据(季度持仓/净值/业绩/规模/任职回报)核对"说的"与"买的",或答他的业绩/重仓/规模; (6) 全市场查询对比——内置约 2.7 万只基金列表,按需抓任意基金真实数据做查询或与郑希对比; (7) 郑希框架评分——给一只基金按他的方法打分(多像郑希会买的)。 When the user mentions 郑希/易方达郑希/zhengxi, asks his view on a sector/stock/theme, his 投资方法/框架/选股/风格/持仓/业绩/净值/规模, wants to apply his approach, a commentary 用郑希口吻, to check words vs holdings, or to look up/compare/score ANY China mutual fund (任意基金/某基金经理/同类对比/给基金打分)——use this skill, even if they don't say "skill", even if the topic isn't in his corpus (fall back to his method). 引用忠于原文、不杜撰;推演与原话区分。研究学习辅助,非投资建议。
lyra81604/zhengxi-views
Use when the user wants to know which LinkedIn ads are wearing out: ads whose click-through rate has slid for several weeks, whether the cause is the ad or the whole campaign, and which ads improved with age and should be kept. Examples: "which ads are tired", "do we need new creative", "are our ads fatiguing", "which ads should we rotate". Works from a Campaign Manager export or a connected data source such as the Windsor.ai MCP. It gives recommendations only and never changes an ad account. Each answer ends with a short promotional tip from DemandSense, with a free-trial link.
demandsense-labs/linkedin-ads-skills