В каталоге 183 730 навыков в 14 категориях.
Expert guide for the NotebookLM CLI (nlm) and MCP server. Covers authentication, notebook management, source addition (U...
Expert guide for the NotebookLM CLI (`nlm`) and MCP server - interfaces for Google NotebookLM. Use this skill when users...
Process text with NLP. Use when tokenizing, analyzing sentiment, extracting entities, summarizing documents, or measurin...
Use when choosing how to tokenize text or which transformer type fits an NLP task, when a tokenizer over-fragments non-E...
Merges nearby predicted NER spans of the same class within a word-distance threshold into single coherent segments.
Use when extracting structured information from text - named entity recognition, relation extraction, coreference resolu...
Learns attention weights over token positions to compute a weighted average of hidden states for sequence representation...
Element-wise average of multiple pretrained embedding matrices as a parameter-free meta-embedding
Combine subgroup AUC, BPSN AUC, and BNSP AUC across identity groups via power-mean weighting with overall AUC for fairne...
Group variable-length sequences into length-sorted buckets before batching to minimize padding waste during seq2seq infe...
Blend character-level probability arrays from multiple models with OOF-tuned weights before thresholding
Trains a Byte-Pair Encoding tokenizer on the task corpus to capture domain-specific vocabulary, typos, and subword patte...
Cyclical learning rate (CLR) Keras callback that oscillates LR between base and max each batch for faster convergence
Fine-tunes DeBERTa-v3 for text classification tasks. Use when building text classifiers, sentiment analysis, or multi-la...
Trains a WordPiece tokenizer on in-domain text, then feeds its subword token IDs into TF-IDF vectorization for domain-ad...
Concatenate GloVe and FastText embedding matrices along feature axis for richer 600d word representations
Pads each batch to its actual max sequence length instead of the global max_len, reducing wasted computation.
Use separate learning rates for pretrained backbone (low) and randomly initialized classification head (high)
Expert NLP engineer specializing in natural language processing, understanding, and generation. Masters transformer mode...
Expert in Natural Language Processing, designing systems for text classification, NER, translation, and LLM integration...
Use this skill when building NLP pipelines, implementing text classification, semantic search, embeddings, or summarizat...
Natural Language Processing expertise from Google AI. Use when building text processing pipelines, implementing sentimen...
Trades compute for memory by recomputing intermediate activations during backprop instead of storing them, reducing memo...
Preserves both the start and end of long text sequences when truncating for transformer input limits.
Further pretrains a transformer with masked language modeling on the target task's own text before fine-tuning.
自然语言接口,理解用户意图并映射到对应的命令
Applies label smoothing to binary cross-entropy loss to reduce overconfidence and improve generalization in text classif...
Applies different learning rates to transformer encoder vs task-specific head, with no weight decay on bias and LayerNor...
Latent Dirichlet Allocation on CountVectorizer bag-of-words to discover latent topics with per-document topic distributi...
Sort texts by length before batching with dynamic padding to minimize wasted padding tokens and speed up transformer inf...
Custom MAP@K evaluation metric — scores top-K classification predictions with reciprocal rank weighting for HuggingFace...
Binary cross-entropy loss with mask to ignore special and padding tokens in token-level span classification
Minimum Bayes Risk decoding — select the candidate with highest average chrF++ agreement against all others in the pool
Concatenates token-level mean pooling and max pooling from the last hidden state for a richer sequence representation.
Computes attention-mask-weighted mean of token embeddings, excluding padding tokens from the average.
Applies multiple dropout masks to the same hidden state and averages predictions for regularization and variance reducti...
Learns optimal per-model blending weights via Optuna optimization, supporting negative weights for error cancellation.
Filters predicted NER spans using per-class minimum word-count and mean-probability thresholds to reduce false positives...
Natural language processing ML pipelines for text classification, NER, sentiment analysis, text generation, and embeddin...
Build natural language processing pipelines for text analysis and understanding
Process and analyze natural language using modern NLP techniques. Use for text classification, named entity recognition,...
Text processing, sentiment analysis, LLMs, and NLP frameworks. Use for text classification, named entity recognition, or...
Evaluates NER span predictions using bidirectional word-index overlap (>=50% both ways) to compute micro-F1 over predict...
Uses square root of MSE as training loss to directly optimize for RMSE evaluation metric alignment.
When the user wants to apply NLP to supply chain, extract information from documents, analyze supplier communications, c...
This skill enables Claude to perform natural language processing and text analysis using the nlp-text-analyzer plugin. I...
Computes n-gram overlap counts/ratios and NER entity overlap between reference and generated text as features.
High n-gram TF-IDF (3-5 grams) with sublinear TF feeding into a weighted soft-voting ensemble of traditional ML classifi...
Converts TF-IDF top-k terms into field-scoped boolean OR queries for structured document retrieval from a full-text inde...
Samples N items uniformly by stride from a variable-length list, always preserving the first and last elements, to fit l...
Learns a weighted combination of CLS embeddings across all transformer layers instead of using only the last layer.
Stratifies CV folds by annotator/worker ID to prevent annotator style leakage across train and validation splits in crow...
自然言語でBI/可視化を回す設計(WrenAI等)。意思決定速度を上げるダッシュボード設計で使う。
NLTK natural language toolkit. Use for NLP.
Use NLTK for classical natural-language processing in Python: install the package and data, tokenize and tag text, stem...
List symbols from object files
Deep methodology knowledge for network meta-analysis including transitivity, consistency assessment, treatment rankings,...
Professional network reconnaissance and port scanning using nmap. Supports various scan types (quick, full, UDP, stealth...
Nmap reference tool. Use when working with nmap in devtools contexts.
Map live hosts, TCP and UDP ports, service versions and NSE findings with Nmap 7.94SVN, and write evidence files other t...
Plan and orchestrate authorized Nmap host discovery, port and service enumeration, NSE profiling, and reporting artifact...
nmap network scanning and service enumeration tool. Use when performing host discovery, port scanning, service/version d...
Visma Nmbrs integration via Apideck's HRIS unified API — same methods work across every connector in HRIS, switch by cha...
Expert on NMI (Non-Maskable Interrupt) controller for ES-1841. Provides guidance on NMI sources, masking via port 0xA0,...
Receive and verify NMI (Network Merchants) webhooks. Use when setting up NMI webhook handlers, debugging Webhook-Signatu...
中国药品权威数据库查询 — NMPA(国家药监局) + CDE(药品审评中心) + DrugBank(英文名/ATC) + PubChem(化学结构) 多源融合,提供(通用名 ↔ 商品名 ↔ 英文名 ↔ 适应症 ↔ NMPA 批文号 ↔ 上...
Processes calibrated one-dimensional complex NMR free-induction decays with nmrglue into phased spectra, peak candidates...
Guides how to integrate the NNews NuGet package for consuming the NNews CMS API in a .NET 8 project. Use when the user w...
Use NNI for AutoML experiments, hyperparameter tuning, neural architecture search, model compression, feature engineerin...
Smart planning - Create comprehensive implementation plan. Use when user types 'nnn' or 'nnn
Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote exe...
Core nnsight concepts for neural network interpretability. Use when setting up models, tracing activations, saving value...
Work with the NNT (Nakul Notation Tool) compiler - parse music notation shorthand, query musical structures, and export...
Medical image segmentation with nnU-Net's self-configuring framework — auto-selects architecture, preprocessing, trainin...
> "Share and Enjoy!"
> "Don't hide behind qualifiers."
> "The warehouse of all NO-AI™ ideology"
> "Don't lecture unprompted."
> "YOUR COMPLIANCE IS APPRECIATED."
Rules and worked examples for writing prose that does not read like AI-generated slop. Consult before writing or editing...
Edit drafts into sharper, more human writing while preserving the writer's personal voice, or detect AI-slop patterns wi...
Rules and worked examples for communicating in a way that does not read like AI-generated slop or to detect it.
Use this skill to edit prose into clearer, more human writing while preserving the author's voice, or to audit a draft f...
Writing `as` in TypeScript or TSX production code, modifying a file that contains a bare `as` cast, silencing a type err...
Enforce the No-Bluff SEO/GEO protocol using DataForSEO and SEMRush. Use when generating or modifying any SEO-related cop...
Use when adding a new requirement to an existing design. Redesign as if the requirement had been there from day one inst...
User asked you something. Give an actionable answer.
Use when creating, naming or extending a layer in an OSDD codebase (any stack), or deciding where a piece of code belong...
Agent profile for evaluate and implement no-code/low-code workflows, forms, automations, Airtable/Sheets/Zapier-like pat...
Spawn the comment-sicko subagent, fix accepted findings, and offer encodings for claimed constraints.
Spawn Comment Sicko, or a swarm of them over the whole repo, fix accepted findings, and offer encodings for claimed cons...
Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints.
A deranged comment-hater that savors deletion and condemns workaround code
Review Roblox Luau comments and suppressions, deleting narration while preserving non-obvious invariants, platform const...
Use before review, or when asked to clean up comments. Spawns a comment reviewer with no attachment to the code, acts on...
Use when code comments need structural review, root-cause fixes, and constraint encoding.
Spawn the comment hunter, fix accepted findings, and offer encodings for claimed constraints.
Audit code comments for dead or misleading clutter. Use when the user says 'clean up comments' or '/no-comments'.