精选 Skills 索引（第 121 页）
按质量分排序的前 8,000 个 Skills，每个都有可收录的独立页面。需要全库 15.8 万条检索时，请使用 Skills 目录。
- sparse-autoencoder-training
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying mo…
- speculative-decoding
Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with li…
- stable-diffusion-image-generation
State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.
- systems-paper-writing
Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides paragraph-level structural blueprints, writing patterns, venue-specific checklists, reviewer guidelines, LaTeX templates, and conferen…
- tensorboard
Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit
- tensorrt-llm
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization…
- torchforge-rl-training
Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.
- training-llms-megatron
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/…
- transformer-lens-interpretability
Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patt…
- verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
- weights-and-biases
Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform
- whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast t…
- ego-browser
ego-browser (ego-lite) is a Chromium-based browser designed from the ground up to be friendly to both human users and AI Agents. AI Agents work in their own isolated space, reusing the user's login state without competing for the browser.…
- canvas
Visual layer of the wiki. Add images, text cards, PDFs, and wiki pages to Obsidian canvas files with auto-positioning inside zones. Integrates with /banana for image capture. Triggers on: /canvas, canvas new, canvas add image, canvas add t…
- defuddle
Strip clutter from web pages before ingesting into the wiki. Removes ads, navigation, headers, footers, and boilerplate: leaving clean readable markdown that saves 40-60% tokens. Triggers on: defuddle, clean this page, strip this url, fetc…
- obsidian-bases
Create and edit Obsidian Bases (.base files): Obsidian's native database layer for dynamic tables, card views, list views, filters, formulas, and summaries over vault notes. Triggers on: create a base, add a base file, obsidian bases, base…
- obsidian-markdown
Write correct Obsidian Flavored Markdown: wikilinks, embeds, callouts, properties, tags, highlights, math, and canvas syntax. Reference this when creating or editing any wiki page. Triggers on: write obsidian note, obsidian syntax, wikilin…
- save
Save the current conversation, answer, or insight into the Obsidian wiki vault as a structured note. Analyzes the chat, determines the right note type, creates frontmatter, files it in the correct wiki folder, and updates index, log, and h…
- think
Apply the 10-principle thinking loop (OBSERVE-OBSERVE-LISTEN-THINK-CONNECT-CONNECT-FEEL-ACCEPT-CREATE-GROW) to any non-trivial problem. Walks Claude through external observation, metacognition, active listening, first-principles analysis,…
- wiki
Claude + Obsidian knowledge companion. Sets up a persistent wiki vault, scaffolds structure from a one-sentence description, and routes to specialized sub-skills. Use for setup, scaffolding, cross-project referencing, and hot cache managem…
- wiki-cli
Default vault-mutation transport for claude-obsidian v1.7+. Wraps the Obsidian CLI (Obsidian 1.12+) as the preferred way to read, write, search, and modify vault notes from Claude — no MCP server, no REST API plugin, no TLS workarounds. Fa…
- wiki-fold
Rollup of wiki log entries into meta-pages. Reads the last 2^k entries from wiki/log.md, writes a structurally-idempotent fold page to wiki/folds/ that links back to children. Extractive summarization (no invention). Dry-run by default, st…
- wiki-mode
Methodology modes for the Compound Vault. Lets the vault declare an organizational style (LYT / PARA / Zettelkasten / Generic) that wiki-ingest, save, and autoresearch consult before filing new pages. Reads `.vault-meta/mode.json`; default…
- wiki-retrieve
Hybrid retrieval primitive for the Compound Vault. Replaces the v1.6 static hot→index→drill read order with contextual-prefix + BM25 + cosine-rerank, modeled on Anthropic's Sept 2024 Contextual Retrieval research (35-49-67% retrieval-failu…
- browser-edge-cases
SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
- hive.chart-creation-foundations
Required reading whenever any chart_* tool is available. Teaches the one-tool embedding contract (call chart_render → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (str…
- hive.colony-progress-tracker
Claim tasks, record step progress, and verify SOP gates in the colony SQLite queue. Applies when your spawn message includes a db_path field.
- hive.context-preservation
Proactively extract critical values from tool results into working notes before automatic context pruning destroys them.
- hive.error-recovery
Follow a structured recovery decision tree when tool calls fail instead of blindly retrying or giving up.
- hive.linkedin-automation
Read before automating LinkedIn with browser_* tools. LinkedIn combines shadow DOM (#interop-outlet), strict Trusted Types CSP that silently drops innerHTML, Lexical composer, native beforeunload dialogs that hang the bridge, and aggressiv…
- hive.note-taking
Maintain a free-form scratchpad of decisions, extracted values, and open questions so context pruning doesn't lose anything you still need.
- hive.quality-monitor
Periodically self-assess output quality to catch degradation before the judge does.
- hive.terminal-tools-foundations
Required reading whenever any shell_* tool is available. Teaches the foreground/background dichotomy (terminal_exec auto-promotes past 30s, returns a job_id you poll with terminal_job_logs), the standard envelope shape (exit_code, stdout,…
- hive.terminal-tools-fs-search
Use terminal_rg / terminal_find when you need raw filesystem search outside the project tree — system configs, /var/log, /etc, archive contents — or when files-tools.search_files is too project-scoped. Teaches the rg vs find vs terminal_ex…
- hive.terminal-tools-job-control
Use when launching anything that runs longer than a minute, anything that streams logs, anything you want to keep running while doing other work — or when terminal_exec auto-backgrounded on you and returned a job_id. Teaches the start→poll…
- hive.terminal-tools-pty-sessions
Use when you need state across calls — building env vars, navigating with cd, driving REPLs (python -i, mysql, psql, node), or responding to interactive prompts (sudo password, ssh host-key confirmation, mysql connection). Teaches the prom…
- hive.terminal-tools-troubleshooting
Read when a terminal-tools call returned something surprising — empty stdout despite no error, exit_code is null, output_handle came back expired, "too many jobs" / "session busy" / "too many PTYs", warning was set unexpectedly, semantic_s…
- hive.writing-hive-skills
Author a new Agent Skill for a Hive agent that conforms to the Agent Skills specification (SKILL.md with YAML frontmatter, optional scripts/references/assets directories). Use when the user asks to create, scaffold, add, or package a new s…
- hive.x-automation
Read before automating X / Twitter with browser_* tools. Verified flows for post, reply, delete, search-and-engage, plus the Draft.js compose quirks that silently disable the send button. Includes the daily-reply and job-market-reply playb…
- web-access
所有联网操作必须通过此 skill 处理，包括：搜索、网页抓取、登录后操作、网络交互等。 触发场景：用户要求搜索信息、查看网页内容、访问需要登录的网站、操作网页界面、抓取社交媒体内容（小红书、微博、推特等）、读取动态渲染页面、以及任何需要真实浏览器环境的网络任务。