“sub-agent”的搜索结果

Runtime router for context decisions during a task — decide read-more / stop / ask / assume / delegate, pick a minimal read-set, and time compaction. Front-page mental model is Write / Select / Compress / Isolate; a lazy pattern index routes to 25 context-engineering patterns, one reference per decision. Use when context is near budget, a long session is degrading, you must choose what to read for a task, or you are deciding /compact vs /clear vs a sub-agent. NOT for structural .claude/ overhead audits (see mk:context-audit); NOT for owning persistence (see mk:memory); NOT for USD cost (see mk:budget).

中文简介正在准备中。当前显示原文说明。

Create a new sub-agent definition file (agents/*.md) for the Claude Code plugin. Use this skill when the user wants to add a new specialized agent, define a new agent type, or create an agent that handles a specific domain of tasks. Trigger whenever the user mentions creating, adding, or defining a new agent or sub-agent.

中文简介正在准备中。当前显示原文说明。

以指定提交、分支或标签为基准，从开发规范和原始需求两个角度审查代码变更。分别报告规范问题与需求缺漏，便于独立判断两方面的结果。

• 审查分支或PR的变更
• 检查指定提交或标签之后的变更
• 对照已记录的开发规范

Set up and operate Ruflo to coordinate multiple AI coding agents. This skill guides project initialization, shared memory, task assignment, health checks, and plugin discovery for work that needs coordinated or parallel execution.

中文简介正在准备中。当前显示英语简介。

• Initializing a multi-agent development environment
• Coordinating parallel work across agents
• Keeping searchable memory across sessions

Turn a music track into a HyperFrames video with text, images, and clips timed to its rhythm and energy. Additional visual assets are optional, and the workflow includes storyboard approval, visual checks, and final MP4 rendering.

中文简介正在准备中。当前显示英语简介。

• Create a promotional video timed to a music track
• Build a slideshow from music and supplied photos
• Make a music video using animated text alone

Prepare Apache Airflow provider releases by classifying pending changes, inspecting pull request diffs for compatibility risks, and applying version bumps and changelog entries. Use existing breeze tooling to regenerate release documentation and consult the release manager on uncertain decisions.

中文简介正在准备中。当前显示英语简介。

• Prepare documentation for a regular provider release cycle
• Classify changes for a selected set of providers
• Review breaking changes and choose version bumps

Work through one difficult mathematics problem with a single agent, saving established arguments and dead ends as progress notes. Resumes from recorded work after interruptions and writes a self-contained proof file, including any unresolved parts.

中文简介正在准备中。当前显示英语简介。

• Work through one difficult mathematics problem
• Resume a proof attempt after an interruption
• Record lemmas, reductions, and dead ends

Work through GitHub pull request feedback by classifying findings, fixing code, running repository checks, and posting replies. With suitable permissions, the skill repeats Copilot reviews until the latest code is reviewed and every thread has an agent response.

中文简介正在准备中。当前显示英语简介。

• Address outstanding Copilot review comments
• Run repeated reviews on a completed pull request
• Explain declined suggestions and hand off design decisions

After explicit opt-in, captures original ideas in the user's exact words and identifies people and companies mentioned in conversations. Stores them in linked knowledge pages with sources, dated facts, and a brief capture log.

中文简介正在准备中。当前显示英语简介。

• Preserving ideas expressed during conversations
• Adding context to people and company pages
• Connecting ideas with related people and concepts

Delegate coding, analysis, planning, and reviews from Codex to Claude Code through its command-line interface. Capture responses, resume sessions, or run independent tasks in parallel, then verify changed files and test results independently.

中文简介正在准备中。当前显示英语简介。

• Delegating implementation or refactoring from Codex
• Getting another model's review of code changes
• Continuing delegated work across multiple sessions

Delegate coding, analysis, refactoring, or reviews from Claude Code to Codex CLI. The skill explains unattended execution, permission settings, output capture, parallel tasks, and independent checks of changed files and test results.

中文简介正在准备中。当前显示英语简介。

• Delegate implementation while retaining planning responsibility
• Get a second opinion on code changes
• Split independent tasks across separate git worktrees

Give subagents or short sessions a compact set of working rules for checking evidence, changing unsuccessful approaches, and verifying results. It also defines agent roles and uses stern corporate-style commentary to push execution forward.

中文简介正在准备中。当前显示英语简介。

• Giving subagents a shared set of execution rules
• Adding evidence checks to short working sessions
• Changing approach after repeated failures

Design tools that agents can select and call correctly. Clarify descriptions, input rules, response formats, and recovery messages, then review overlapping tools and naming conventions using examples of successful and failed calls.

中文简介正在准备中。当前显示英语简介。

• Writing descriptions and input rules for new tools
• Investigating incorrect tool selection and calls
• Improving error messages with recovery steps

Write or refine Chinese spoken video scripts about AI technology, interview questions, and current topics. New scripts follow research and outline approval, while revisions provide specific wording changes and reasons for the edits.

中文简介正在准备中。当前显示英语简介。

• Writing a spoken script explaining AI technology
• Turning technical interview questions into video scripts
• Preparing a script about a current AI topic

Independently checks code changes against the original issue, runs tests, and verifies that edits stay within scope. Produces an approval, rejection, or request for human judgment, supported by test results and specific findings.

中文简介正在准备中。当前显示英语简介。

• Independently reviewing another agent's implementation
• Checking that a CI failure was actually fixed
• Finding unrelated edits or disabled tests

Review code changes with an independent agent from a different model provider. After tests pass, check a fixed diff against acceptance criteria, send blocking issues back to the implementer, and repeat until review passes.

中文简介正在准备中。当前显示英语简介。

• Get an independent review from another model provider
• Check a pull request against acceptance criteria
• Turn blocking findings into concrete fix tasks

Test polly's coordination of coding agents using a scripted mock language model and live command-line tools. Check startup, task dispatch, safety rules, parallel delegation, and whether real agents make appropriate orchestration decisions.

中文简介正在准备中。当前显示英语简介。

• Reproducing polly orchestration bugs
• Checking safety rules with a mock language model
• Verifying delegation and cross-review with real agents

Test Omnigent’s GitHub Copilot SDK integration on a live local server. Build test agents, run real conversations, and inspect authentication, model selection, tool connections, policy enforcement, and leftover processes to investigate integration failures.

中文简介正在准备中。当前显示英语简介。

• Checking Copilot authentication with a real conversation
• Testing model selection and tool connections
• Verifying sub-agent dispatch and policy enforcement

Run independent development tasks in parallel, each with a separate git worktree and implementation agent. Each agent opens its own pull request, which receives cross-review before being handed to a human for review and merging.

中文简介正在准备中。当前显示英语简介。

• Implementing independent changes in parallel
• Assigning separate working directories to agents
• Preparing an individual pull request for each task

Test Omnigent’s Cursor SDK integration against a live local server. Build test agents, run real conversations, and check authentication, model selection, and tool connections while investigating empty responses or bridge processes left after shutdown.

中文简介正在准备中。当前显示英语简介。

• Test Cursor authentication and real responses
• Verify file operations and sub-agent delegation
• Investigate invalid models and empty output

Test Omnigent’s Antigravity SDK integration against a live local server. Build test agents and run real conversations to check Gemini authentication, model selection, tool connections, and failures across the full execution path.

中文简介正在准备中。当前显示英语简介。

• Test authentication and real Gemini responses
• Verify tool connections and sub-agent delegation
• Investigate model selection failures

Deeply analyze any empirical economics PDF using the five-question framework (五问框架): research question, identification strategy, core estimand, robustness logic, and scholarly contribution. Converts the PDF to Markdown, dispatches 5 parallel sub-agents (one per question), then compiles a 2000–3000 Chinese-character report with a transferable methodology checklist. Optimized for empirical economics papers in top journals (AER, QJE, JPE, RFS, RoF, JFE, etc.). Use this skill whenever the user wants to: - systematically analyze an empirical economics paper - apply the 五问框架 (five-question framework) to a paper - extract transferable research design elements from a top-journal paper - understand a paper's identification strategy, estimand, or robustness logic - run /five-questions, /five-q, or /paper-five-q-analysis - "帮我做五问分析", "用五问框架分析这篇论文", "analyze this paper with the five-question framework"

中文简介正在准备中。当前显示原文说明。

Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".

中文简介正在准备中。当前显示原文说明。

Plan and execute large refactor or rewrite efforts efficiently with parallel multi-agent analysis and implementation. Use when a user asks to refactor many files, split workstreams, analyze a target code area, and coordinate sub-agents with clear ownership and dependency-aware execution.

中文简介正在准备中。当前显示原文说明。