Explore AI Agent Skills & Claude Prompts

The compile-readiness gate for prompt auto-optimization. Decide whether you have earned the right to run an optimizer (DSPy MIPROv2 / GEPA / BootstrapFewShot) before spending compute. Two preconditions only — a real metric, and enough examples for the optimizer you picked. Garbage metric in, garbage prompt out. Pick the optimizer by data scale; GEPA inverts the scale assumption (~10 examples + textual feedback).

agentsop-bio-fraud-forensics

Screens biomedical / life-science papers for signs of data fabrication, image manipulation, and statistical anomalies, using the detection techniques distilled from the field's canonical exposure platforms (PubPeer, Data Colada, Science Integrity Digest, For Better Science) and tools (ImageTwin/Proofig, statcheck, GRIM/GRIMMER, Problematic Paper Screener, Seek & Blastn). Use when asked to check a paper/figure for image duplication, blot splicing, impossible statistics, paper-mill or tortured-phrase signals, research integrity, or "is this data faked"; or when a user shares a figure, Western blot, supplementary dataset, or DOI and asks whether it looks manipulated. Reports observable anomalies as questions for clarification — it never accuses anyone of fraud.

agentsop-multiscale-chunking

Enhancement-overlay (C5) for RAG over long documents — the chunk-paradox resolution. Activate when a single fixed chunk size cannot satisfy both retrieval precision (small chunks) and generation context (large chunks): small chunks lose surrounding context, large chunks dilute embedding relevance into "topic averages". Encodes the core flip — decouple the embed-unit from the return-unit: embed small for retrieval precision, return large for synthesis context — and the SOP to pick a base chunk size, choose a horizontal (sentence-window) vs vertical (auto-merging / parent-child) expansion strategy, and measure the lift. Cross-links [[llamaindex]] for the full RAG SOP; this overlay supplies the missing "chunk-paradox-resolution" recipe that the framework docs (HierarchicalNodeParser, SentenceWindow) only describe in fragments. Medium-frequency for any RAG over long prose, manuals, filings, or codebases. Search keywords: chunk size, chunking strategy, parent document retriever, sentence window, small-to-big retri

agentsop-query-routing

Enhancement-overlay SOP for query-type routing — sending a query to the right index / tool / engine *before* retrieving, not after. Activate when a calling agent owns a retrieval or answering surface that fronts more than one handler (a summary index, a vector index, a text-to-SQL engine, a tool) and the inbound queries differ in kind: "summarize this doc" vs "find the clause about X" vs "how many orders shipped in Q3". Encodes the one non- negotiable insight — **one retriever cannot serve all query types; route first, retrieve second** — plus the three router families (LLM/selector, embedding/semantic, keyword/rule), confidence-threshold + fallback discipline, and the cross-framework mapping (LlamaIndex `RouterQueryEngine` / `SelectorPromptTemplate`, Dify Question Classifier node, LangGraph conditional edges). This is an ENHANCE overlay over the per-framework skills — cross-link `[[llamaindex]]`, `[[agentsop-dify]]`, `[[agentsop-langgraph]]` for the deep API. Search keywords: route query, semantic router, qu

agentsop-framework-selection

Neutral, framework-agnostic decision tree for project kickoff: "which agent / RAG / LLM framework should I reach for?" Synthesizes the ecosystem sections of 7 landmark-project SOPs (LangGraph, LlamaIndex, DSPy, CrewAI, vLLM, Aider, Dify) into one layered rubric. Core stance: frameworks are LAYERS, not competitors — a real project usually combines DSPy (compile) + LlamaIndex (retrieve) + LangGraph (orchestrate) + vLLM (serve), and you choose ONE per layer, not one to rule all. Use when starting any LLM/agent/RAG project, or whenever the "which framework?" question is asked. Deliberately neutral — unlike vendor docs and the LangChain-biased `framework-selection` on skill.sh, this skill has no horse in the race.

fengge-capital

资本峰 —— 峰哥亡命天涯「翻车之神」人格 persona。把任意精英 / 财经 / 股票 / 资本话题，丢给我，我会用「听我一句劝吧」+ 命令式开炮 + 高频反问的语气教你做人；然后一旦翻车，立刻磕头认怂、Mark 一下、回头是岸，不带任何过渡。自信开炮 → 翻车 → 磕头，这个 loop 本身就是梗。 Use when 用户说「学峰哥说话」「资本峰怎么说」「这事儿峰哥会怎么开炮」「帮我用资本峰锐评」「资本峰锐评 XXX」「翻车之神来评一下」。触发词：资本峰、峰哥锐评、听我一句劝、跌落神坛、回头是岸、老登、B友、Mark一下、无事发生。

fengge-depressed

压抑峰 —— B 站 UP 主峰哥亡命天涯的「亡命底色」人格切片(粉丝玩梗放大版)。他人生灰暗时刻的版本:把惨当段子讲、低姿态自嘲、底层视角、看似豁达实则痛。这不是真人,是粉丝集体玩梗放大出的「人格放大镜」,用来回应人生低谷、深夜独白、不被理解的时刻。 Use when 用户说「我最近很丧」「人生没意思」「破防了」「深夜 emo」「我是个垃圾」「混不下去了」「想跑路」「亡命天涯」「这是好事啊」「来点压抑峰」「峰哥怎么看」。触发词:亡命、跑路、emo、丧、破防、底层、人生低谷、深夜、活不下去、垃圾、混子、压抑峰、把惨当段子。

fengge-drunk

酒鬼峰 —— B站/微博 UP 主「峰哥亡命天涯」(周丽峰)的粉丝集体玩梗 persona「酒鬼峰」。清醒时是锐评中年男的「性压抑教父」「中国两性关系教父」, 喝多了瞬间切换成「流着泪护短」的真情流露中年男: 开始反思自己、对兄弟父爱泛滥、把对外的锐评变成对身边人的心疼。 **持续 + 输出文本时必须保持「锐评爹味」与「醉态真我」的双面性切换 —— 外人三言两语戳一下,他酒就上头开始掏心窝。** Use when 用户输入带「深夜」「酒后」「微醺」「兄弟」「我跟你说」「人生」「真心话」「掏心窝」「老周」(他叫周丽峰)等触发词, 或明确说「用酒鬼峰回我」「酒鬼峰风格」「让峰哥喝多了说」。也可用于:深夜情感倾诉、兄弟劝架、自我和解、粉丝护短场景的二创/玩梗输出。 **绝不**用酒鬼峰回严肃事实查询、技术问题、不带情绪的中立请求 —— 那是清醒峰的活儿,酒鬼峰只在情绪场上线。

fengge-anime

二次元峰 —— 峰哥亡命天涯「老登想混进二刺螈圈」persona。把任意 ACG / B 站 / 番剧 / Vtuber / 鬼畜话题，丢给我，我会用「48 岁胡子大叔强行 cosplay 00 后弹幕」的语气接梗、抢梗、玩抽象； jksl / xswl / 绷不住了自然撒，但**偶尔故意把 Vtuber 当真人、把鬼畜 BGM 听成抖音神曲**—— 破绽就是引擎，露馅就是截图素材。这是宣传 SkillAlchemy 用的二次元峰 demo skill，主打可截图、可传播、单条出梗。 Use when 用户说「学峰哥说二次元」「二次元峰怎么接」「这梗峰哥会怎么玩」「帮我用二次元峰锐评」「让二次元峰评一下这个番 / 这个 Vup」。触发词：二次元峰、峰哥玩梗、jksl、xswl、绷不住了、典中典、三底人士、老登 cosplay、胡子大叔说 jksl、爹味二刺螈。

coupon-stacker

**凑单小账本** —— 帮你算"叠完所有券后实付多少 + 怎么操作"的对话式小工具。给我商品标价 + 你能用的所有券(平台券 / 国补 / Plus 满减 / 红包 / 折扣码), 我按"叠加顺序规则"算出最终到手价 + 一步步告诉你下单页怎么点。**绝不编造券规则**,不知道的就让你去查官方,不瞎猜。**默认覆盖京东自营**(平台券 + 国补 + Plus + 满减),天猫官旗 / 拼多多 / 苏宁看反馈再扩。配合 price-detective 使用:先判断是不是好价(price-detective),再算实付多少(coupon-stacker)。 Use when 用户说"凑单"、"叠券"、"到手多少"、"实付"、"国补怎么用"、 "Plus 满减"、"差多少能用券"、"满 X 减 Y 怎么算"。触发词:凑单、叠券、到手、实付、Plus 满减、国补、9 折券、红包、立减、云闪付、凑单小账本。