Skills Plugins MCP Prompt Model 博客 我的中心
生活与工具 #research #agent

plan-update

Use when you need to update a plan document with deep research across all compiler passes. Launches parallel subagents to analyze how a topic affects every compiler phase, then consolidates findings into the plan doc.

DeepseekModel 官方收录技能 质量 优秀 · 90 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=react-react-compiler-claude-skills-plan-update-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name plan-update description Use when you need to update a plan document with deep research across all compiler passes. Launches parallel subagents to analyze how a topic affects every compiler phase, then consolidates findings into the plan doc. Plan Update Deep-research a topic across all compiler passes and update a plan document. Arguments: $ARGUMENTS: <plan-doc-path> <topic/question> Example: compiler/docs/rust-port/rust-port-0001-babel-ast.md scope resolution strategy Example: compiler/docs/rust-port/rust-port-architecture.md error handling patterns Instructions Step 1: Read context Read these files to understand the current state: The plan doc specified in $ARGUMENTS compiler/docs/rust-port/rust-port-architecture.md (architecture guide and port conventions) compiler/packages/babel-plugin-react-compiler/docs/passes/README.md (pass overview) Step 2: Launch parallel analysis agents Launch 8 parallel Agent tool calls using the analyze-pass-impact agent. Each agent analyzes one phase group. Pass each agent the topic from $ARGUMENTS and the list of pass doc files for its phase. Phase groups and their pass docs: Lowering & SSA (passes 01-03): 01-lower.md , 02-enterSSA.md , 03-eliminateRedundantPhi.md Optimization & Types (passes 04-06): 04-constantPropagation.md , 05-deadCodeElimination.md , 06-inferTypes.md Function & Effect Analysis (passes 07-09): 07-analyseFunctions.md , 08-inferMutationAliasingEffects.md , 09-inferMutationAliasingRanges.md Reactivity & Scope Variables (passes 10-14): 10-inferReactivePlaces.md , 11-inferReactiveScopeVariables.md , 12-rewriteInstructionKindsBasedOnReassignment.md , 13-alignMethodCallScopes.md , 14-alignObjectMethodScopes.md Scope Alignment & Terminals (passes 15-20): 15-alignReactiveScopesToBlockScopesHIR.md , 16-mergeOverlappingReactiveScopesHIR.md , 17-buildReactiveScopeTerminalsHIR.md , 18-flattenReactiveLoopsHIR.md , 19-flattenScopesWithHooksOrUseHIR.md , 20-propagateScopeDependenciesHIR.md Reactive Function & Transforms (passes 21-30): 21-buildReactiveFunction.md , 22-pruneUnusedLabels.md , 23-pruneNonEscapingScopes.md , 24-pruneNonReactiveDependencies.md , 25-pruneUnusedScopes.md , 26-mergeReactiveScopesThatInvalidateTogether.md , 27-pruneAlwaysInvalidatingScopes.md , 28-propagateEarlyReturns.md , 29-promoteUsedTemporaries.md , 30-renameVariables.md Codegen & Optimization (passes 31, 34-38): 31-codegenReactiveFunction.md , 34-optimizePropsMethodCalls.md , 35-optimizeForSSR.md , 36-outlineJSX.md , 37-outlineFunctions.md , 38-memoizeFbtAndMacroOperandsInSameScope.md Validation Passes (passes 39-55): 39-validateContextVariableLValues.md , 40-validateUseMemo.md , 41-validateHooksUsage.md , 42-validateNoCapitalizedCalls.md , 43-validateLocalsNotReassignedAfterRender.md , 44-validateNoSetStateInRender.md , 45-validateNoDerivedComputationsInEffects.md , 46-validateNoSetStateInEffects.md , 47-validateNoJSXInTryStatement.md , 48-validateNoImpureValuesInRender.md , 49-validateNoRefAccessInRender.md , 50-validateNoFreezingKnownMutableFunctions.md , 51-validateExhaustiveDependencies.md , 53-validatePreservedManualMemoization.md , 54-validateStaticComponents.md , 55-validateSourceLocations.md Each agent prompt should be: Analyze how the topic "<topic>" affects the following compiler passes. Read each pass's documentation in compiler/packages/babel-plugin-react-compiler/docs/passes/ and its implementation source. Also read compiler/docs/rust-port/rust-port-architecture.md for port conventions. Pass docs to analyze: <list of pass doc filenames> For each pass, report: - Pass name and purpose (1 line) - Impact: "none", "minor" (mechanical changes only), "moderate" (logic changes), or "significant" (redesign needed) - If impact is not "none": specific details of what changes are needed - Key code locations in the implementation (file:line references) Be thorough but concise. Focus on concrete impacts, not speculation. Step 3: Consolidate and update After all agents complete: Merge their findings into a coherent analysis Group by impact level (significant > moderate > minor > none) Update the plan document. The final state should reflect the latest findings and understanding: Update the plan document in-place to remove outdated content and describe the latest understanding. KEEP any existing content that is still relevant REMOVE outdated or now-irrelevant content Per-pass impact summary table (updated in place, ie update each section based on new findings) Detailed notes for passes with moderate+ impact Updated "Current status" or "Remaining Work" section if applicable Step 4: Show summary Show the user a brief summary of findings: how many passes are affected at each level, and the key insights.
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。