166 个 · 按相关度排序
Investigate why a PyTorch model behaves a certain way by changing its internal values with pyvene. The skill guides experiments that swap or remove activations, train interventions, compare outputs, and save or share intervention configurations.
中文简介正在准备中。当前显示英语简介。
Investigate how transformer language models produce predictions with TransformerLens. This skill guides inspection of intermediate values and attention patterns, plus experiments that replace internal values to test their effects and identify components involved in model behavior.
中文简介正在准备中。当前显示英语简介。
Investigate how transformer language models produce predictions with TransformerLens. This skill guides inspection of intermediate values and attention patterns, plus experiments that replace internal values to test their effects and identify components involved in model behavior.
中文简介正在准备中。当前显示英语简介。
Guides experiments that change internal values in PyTorch models to test what causes their behavior. Uses pyvene to swap or remove activations, locate influential components, train interventions, and save experiments for reuse.
中文简介正在准备中。当前显示英语简介。
Investigate how transformer language models produce predictions with TransformerLens. This skill guides inspection of intermediate values and attention patterns, plus experiments that replace internal values to test their effects and identify components involved in model behavior.
中文简介正在准备中。当前显示英语简介。
Plan, implement, or review dispatching parallel agents work in an existing codebase with compatibility, security, and verification controls. Use when the user explicitly requests dispatching parallel agents work.
中文简介正在准备中。当前显示原文说明。
通过 Claude DevFleet 将开发工作分配给多个 Claude Code 智能体，在独立工作目录中并行执行，并管理任务依赖、进度以及变更、测试和错误报告。
Implement Azure Cosmos DB document operations in Rust using its NoSQL SDK. The skill covers authentication, database and container clients, creating and reading records, replacing or patching fields, and deleting items with partition keys.
中文简介正在准备中。当前显示英语简介。
Inspect executable files from the command line with radare2, starting with file details, strings, and imported functions. Trace code through disassembly, compare binaries, automate inspections, or apply explicitly requested patches with verification.
中文简介正在准备中。当前显示英语简介。
Dispatches due Xiaohongshu tasks already locked in the operations dashboard to their assigned account workers. It checks task identifiers, records delivery failures, and continues dispatching other tasks without operating the browser or changing content.
中文简介正在准备中。当前显示英语简介。
End-to-end methodology for reproducing AND patching publicly disclosed, already-fixed vulnerabilities: sourcing the advisory and the fix commit, reading the patch diff, rebuilding a pinned offline lab, building a vulnerable-vs-patched contrast harness with proper controls, minimizing the trigger, designing and validating the fix, and writing it up as a blog post or whitepaper. Use this skill whenever the user wants to reproduce, demonstrate, understand, teach, or write about a known vulnerability or security incident: "reproduce this CVE", "build a repro for", "how did EchoLeak / Log4Shell / the postmark-mcp package actually work", "show me how this exploit worked", "write a blog post about this vulnerability", "set up a lab for CVE-XXXX", "pull the NVD/CVE/OSV details", or "make a PoC I can publish". Use it just as eagerly for the patching half: "how should this be fixed", "is this patch correct", "what would have caught this", "review this security fix", "write a regression test for this CVE", "how do I backport this", or "what's the right fix for this bug class" - even when the user never says the word "reproduce". Reach for it on vague asks like "help me understand this security advisory", "I want to write about this incident", or "can we test whether we were vulnerable" when a disclosed vulnerability is in play. This skill only handles ALREADY-PATCHED, publicly disclosed issues and refuses zero-day discovery, live targets, and weaponization.
中文简介正在准备中。当前显示原文说明。
Runs servers and corporate systems — patching, configuration baselines, change control, capacity, and the routine that prevents incidents. Use this to establish a patching cadence, standardize server configuration, plan a maintenance window, decide change control for infrastructure, or clean up systems that have drifted from any known state.
中文简介正在准备中。当前显示原文说明。
Use when legacy prompts or older framework flows reference `dispatching-parallel-agents` and you need the modern runtime to resolve that capability cleanly.
中文简介正在准备中。当前显示原文说明。
Bypass the Windows Antimalware Scan Interface (AMSI) using memory patching, reflection, and obfuscation techniques. Execute undetected PowerShell, VBScript, JScript, and .NET assemblies in-memory without triggering Microsoft Defender or third-party AV/EDR solutions. Use this skill during Red Team engagements when loading offensive tools (Mimikatz, Rubeus, SharpHound) in memory on defended Windows endpoints.
中文简介正在准备中。当前显示原文说明。
Fornece orientação para realizar intervenções causais em modelos PyTorch usando o framework de intervenção declarativa do pyvene. Use ao conduzir rastreamento causal, activation patching, treinamento de intervenção de intercâmbio ou testar hipóteses causais sobre o comportamento do modelo.
中文简介正在准备中。当前显示原文说明。
Fornece orientação para pesquisa de interpretabilidade mecanística usando TransformerLens para inspecionar e manipular internals de transformers via HookPoints e caching de ativações. Use ao fazer engenharia reversa de algoritmos de modelos, estudar padrões de atenção ou realizar experimentos de activation patching.
中文简介正在准备中。当前显示原文说明。
Use multiple Claude agents to investigate and fix independent problems concurrently
中文简介正在准备中。当前显示原文说明。
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 patterns, or performing activation patching experiments.
中文简介正在准备中。当前显示原文说明。
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.
中文简介正在准备中。当前显示原文说明。
将相互独立的缺陷或测试失败分配给不同代理，并行调查和修复。明确各自范围与报告要求，再检查修改冲突、运行完整测试并整合结果。
检查Paperclip实例创建的拉取请求，说明用途及合并准备情况，将未就绪的请求转交修复，并根据最新变更的验证结果报告状态。
Consult a field reference for slide JSON, the structured data behind each page. It explains where content and styling live, how edits are validated, and why a saved change may display differently or have no visible effect.
中文简介正在准备中。当前显示英语简介。
Guide slide edits with concrete rules for text size, spacing, alignment, and color contrast. Use it when replacing content or adjusting individual elements to keep pages readable and consistent with their existing design.
中文简介正在准备中。当前显示英语简介。
Find and edit specific fields in OpenMAIC classroom documents. This skill maps pages, content, and playback actions to their data locations, guides small changes, and checks saved results, with references for quizzes, interactive content, and projects.
中文简介正在准备中。当前显示英语简介。