[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-skill-langchain-ai-deep-agents-orchestration-ru":3,"guides-for-langchain-ai-deep-agents-orchestration":260,"similar-k17e86mrthw9p3ka6b379d4nfs867gae":261},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":22,"identity":187,"isFallback":192,"parentExtension":193,"providers":238,"relations":241,"repo":242,"workflow":259},1778053782268.1978,"k17e86mrthw9p3ka6b379d4nfs867gae",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":12,"sourceUrl":13,"tags":14},"INVOKE THIS SKILL when using subagents, task planning, or human approval in Deep Agents. Covers SubAgentMiddleware, TodoList for planning, and HITL interrupts.",{},"Deep Agents Orchestration","https://github.com/langchain-ai/langchain-skills/tree/HEAD/config/skills/deep-agents-orchestration",[15,16,17,18,19,20,21],"deep-agents","orchestration","subagents","task-planning","human-in-the-loop","langchain","langgraph",{"_creationTime":23,"_id":24,"extensionId":5,"locale":25,"result":26,"trustSignals":176,"workflow":185},1778053822111.3547,"kn740gdyzssnqqa87k91v8rhmn866mks","en",{"checks":27,"evaluatedAt":166,"extensionSummary":167,"promptVersionExtension":168,"promptVersionScoring":169,"rationale":170,"score":171,"summary":172,"tags":173,"targetMarket":174,"tier":175},[28,33,36,39,43,46,50,55,58,61,65,70,73,77,80,83,86,89,92,95,99,103,107,111,115,118,121,124,128,131,134,137,140,143,147,150,153,156,159,163],{"category":29,"check":30,"severity":31,"summary":32},"Practical Utility","Problem relevance","pass","The description clearly states that the skill is for using subagents, task planning, or human approval in Deep Agents, addressing specific components like SubAgentMiddleware, TodoList, and HITL interrupts.",{"category":29,"check":34,"severity":31,"summary":35},"Unique selling proposition","The skill implements distinct middleware components for Deep Agents (SubAgentMiddleware, TodoListMiddleware, HumanInTheLoopMiddleware) which go beyond basic agent capabilities by offering structured task delegation, planning, and approval workflows.",{"category":29,"check":37,"severity":31,"summary":38},"Production readiness","The skill covers the full lifecycle for its described features, including setup, configuration, and workflow examples for subagents, task planning, and human-in-the-loop approval, indicating readiness for use in production workflows.",{"category":40,"check":41,"severity":31,"summary":42},"Scope","Single responsibility principle","The skill focuses on the orchestration capabilities within Deep Agents (subagents, task planning, human approval) and does not extend into unrelated domains.",{"category":40,"check":44,"severity":31,"summary":45},"Description quality","The displayed description accurately and concisely reflects the skill's capabilities, mentioning its use with subagents, task planning, and human approval in Deep Agents.",{"category":47,"check":48,"severity":31,"summary":49},"Invocation","Scoped tools","The skill utilizes specific tools like `task`, `write_todos`, and interrupt mechanisms for `deploy_to_prod`, `execute_sql`, etc., which are scoped to their respective functionalities.",{"category":51,"check":52,"severity":53,"summary":54},"Documentation","Configuration & parameter reference","info","While configuration options like `subagents`, `tools`, `interrupt_on`, and `checkpointer` are mentioned and shown in examples, explicit defaults for all parameters and a clear precedence order for configuration are not detailed.",{"category":40,"check":56,"severity":31,"summary":57},"Tool naming","Tools like `task`, `write_todos`, `run_tests`, and `deploy_to_prod` are descriptively named and follow conventions.",{"category":40,"check":59,"severity":31,"summary":60},"Minimal I/O surface","The input schemas for tools and configurations are well-defined (e.g., list of dicts for todos, schema objects for subagents), and the responses are specific to the action performed, avoiding unnecessary dumps.",{"category":62,"check":63,"severity":53,"summary":64},"License","License usability","The README mentions the project is in early development and points to a separate repository for skills, but no explicit LICENSE file or SPDX identifier is directly present in this skill's directory. The overall project's license is not clearly stated here.",{"category":66,"check":67,"severity":68,"summary":69},"Maintenance","Commit recency","not_applicable","No commit history is available for this specific skill's directory.",{"category":66,"check":71,"severity":31,"summary":72},"Dependency Management","The skill relies on `langchain`, `langgraph`, and `deepagents`, which are likely managed through standard package managers (npm/yarn for TS, pip for Python). The README indicates this is part of a larger project, implying dependency management is handled at the project level.",{"category":74,"check":75,"severity":31,"summary":76},"Security","Secret Management","The skill's code examples reference API keys via environment variables (`OPENAI_API_KEY`, `ANTHROPIC_API_KEY`) and do not expose secrets directly in the code or in tool outputs.",{"category":74,"check":78,"severity":31,"summary":79},"Injection","The skill primarily interacts with LangChain/LangGraph components and does not appear to load or execute arbitrary third-party data as instructions.",{"category":74,"check":81,"severity":31,"summary":82},"Transitive Supply-Chain Grenades","The skill relies on established libraries and does not fetch remote code or data at runtime for execution. All logic appears to be contained within the bundled code and its dependencies.",{"category":74,"check":84,"severity":31,"summary":85},"Sandbox Isolation","The skill operates within the framework's sandbox and does not show any attempts to access or modify files outside its designated scope. Paths are handled by the framework.",{"category":74,"check":87,"severity":31,"summary":88},"Sandbox escape primitives","No detached-process spawns or deny-retry loops are evident in the provided skill code.",{"category":74,"check":90,"severity":31,"summary":91},"Data Exfiltration","The skill's examples show API keys being managed via environment variables and do not contain any instructions or code paths that would exfiltrate confidential data to third parties.",{"category":74,"check":93,"severity":31,"summary":94},"Hidden Text Tricks","The bundled files (SKILL.md) are free of hidden-steering tricks such as HTML comments, invisible Unicode characters, or bidi overrides.",{"category":96,"check":97,"severity":31,"summary":98},"Hooks","Opaque code execution","The skill code is presented in clear, readable source format (Python and TypeScript examples) without obfuscation, base64 payloads, or runtime code fetching.",{"category":100,"check":101,"severity":31,"summary":102},"Portability","Structural Assumption","The skill relies on the Deep Agents framework and standard LangChain/LangGraph constructs, making no assumptions about user-specific project directory structures outside of what the framework manages.",{"category":104,"check":105,"severity":68,"summary":106},"Trust","Issues Attention","Issue tracking data (opened/closed) is not available for this specific skill's directory.",{"category":108,"check":109,"severity":53,"summary":110},"Versioning","Release Management","The skill is part of a larger project ('langchain-skills') and the README mentions versioning, but no specific version number is declared for this skill itself within its SKILL.md or associated manifests, and installation instructions point to the main branch.",{"category":112,"check":113,"severity":53,"summary":114},"Code Execution","Validation","While the code examples demonstrate structured inputs for tools and configurations (e.g., `zod` for TypeScript), explicit mention of a schema library for all input validation and sanitization across all tools is not detailed in the SKILL.md.",{"category":74,"check":116,"severity":31,"summary":117},"Unguarded Destructive Operations","Potential destructive operations like file writes or SQL execution are guarded by Human-in-the-Loop approval mechanisms, requiring explicit user confirmation.",{"category":112,"check":119,"severity":31,"summary":120},"Error Handling","The examples show proper error handling for interrupts and rejections, with clear feedback messages and the ability to resume execution, indicating a fail-closed behavior for unhandled states.",{"category":112,"check":122,"severity":68,"summary":123},"Logging","The skill focuses on agent orchestration logic. Logging of destructive actions or outbound calls to a local audit file is not explicitly described as a feature of this specific skill, but is likely handled by the underlying framework (Deep Agents).",{"category":125,"check":126,"severity":31,"summary":127},"Compliance","GDPR","The skill itself does not appear to handle personal data directly; it orchestrates agent actions. Any personal data processing would be within the agents or tools it delegates to, and the examples do not suggest direct submission of personal data to third parties.",{"category":125,"check":129,"severity":31,"summary":130},"Target market","The skill's functionalities (subagents, task planning, HITL) are general-purpose and not tied to any specific geographic or legal jurisdiction. It is designed to work globally.",{"category":100,"check":132,"severity":31,"summary":133},"Runtime stability","The skill is designed to work within the Deep Agents framework and LangChain/LangGraph ecosystems, with examples provided for both Python and TypeScript, suggesting multi-platform compatibility. It does not appear to rely on specific OS or shell features.",{"category":47,"check":135,"severity":31,"summary":136},"Precise Purpose","The description clearly states the skill's purpose for subagents, task planning, and human approval in Deep Agents, naming the specific middleware components and their functions.",{"category":47,"check":138,"severity":31,"summary":139},"Concise Frontmatter","The frontmatter is concise, clearly stating the skill's purpose and the components it covers (SubAgentMiddleware, TodoList, HITL interrupts) within the character limit.",{"category":51,"check":141,"severity":31,"summary":142},"Concise Body","The SKILL.md content is well-structured with clear sections and code examples, and it effectively delegates deeper material to conceptual explanations and code snippets rather than embedding large amounts of text.",{"category":144,"check":145,"severity":31,"summary":146},"Context","Progressive Disclosure","The SKILL.md uses clear sections and code blocks for examples, effectively disclosing information progressively. Complex procedures are explained conceptually rather than embedding large data blobs.",{"category":144,"check":148,"severity":68,"summary":149},"Forked exploration","This skill focuses on orchestrating agent behavior rather than deep code exploration or research, so the `context: fork` setting is not applicable.",{"category":29,"check":151,"severity":31,"summary":152},"Usage examples","The skill provides clear, end-to-end Python and TypeScript examples for using subagents, todo lists, and human-in-the-loop approval, demonstrating invocation and expected outcomes.",{"category":29,"check":154,"severity":31,"summary":155},"Edge cases","The skill documentation addresses limitations and failure modes, such as the statelessness of subagents, the requirement for `thread_id` for persistence, and the need for a `checkpointer` with interrupts, along with recovery steps.",{"category":112,"check":157,"severity":68,"summary":158},"Tool Fallback","The skill relies on core Deep Agents and LangChain tools which are assumed to be available. There are no external custom tools mentioned that would require fallbacks or specific versioning declarations.",{"category":160,"check":161,"severity":31,"summary":162},"Safety","Halt on unexpected state","The skill's design for interrupts and rejections implies a fail-closed mechanism, halting execution on unexpected states and allowing explicit user decisions (approve, reject, edit) to resume, with clear reporting.",{"category":100,"check":164,"severity":31,"summary":165},"Cross-skill coupling","The skill focuses on Deep Agents orchestration and does not appear to implicitly rely on other specific skills being loaded. Cross-skill coordination is not a feature of this skill.",1778053799678,"This skill provides middleware for Deep Agents, enabling task delegation to specialized subagents, structured task planning via a TodoList, and human-in-the-loop approval for sensitive operations. It leverages LangChain and LangGraph components for its functionality.","2.0.0","3.4.0","This skill demonstrates strong adherence to best practices across documentation, security, and code execution. It provides clear examples, robust error handling for its orchestration features, and handles sensitive operations with human-in-the-loop approval. The 'Release Management' check is rated 'info' due to the lack of a specific version for this skill within its manifest.",95,"A high-quality skill for orchestrating complex agent workflows in Deep Agents, offering robust features for subagents, task planning, and human approval.",[15,16,17,18,19,20,21],"global","verified",{"codeQuality":177,"collectedAt":178,"documentation":179,"maintenance":181,"security":182,"testCoverage":184},{},1778053784683,{"descriptionLength":180,"readmeSize":8},159,{},{"hasNpmPackage":183,"smitheryVerified":183},false,{"hasCi":183,"hasTests":183},{"updatedAt":186},1778053822111,{"githubOwner":188,"githubRepo":189,"locale":25,"slug":190,"type":191},"langchain-ai","langchain-skills","deep-agents-orchestration","skill",true,{"_creationTime":194,"_id":195,"community":196,"display":197,"identity":207,"parentExtension":209,"providers":232,"relations":236,"workflow":237},1778053782268.1963,"k1710b29xhgdykchws1tmnjrnx866nxv",{"reviewCount":8},{"description":198,"installMethods":199,"name":200,"sourceUrl":201,"tags":202},"Agent skills for building agents with LangChain, LangGraph, and Deep Agents",{},"LangChain Skills","https://github.com/langchain-ai/langchain-skills",[20,21,15,203,204,205,206],"ai-agents","rag","skills","plugins",{"githubOwner":188,"githubRepo":189,"locale":25,"slug":189,"type":208},"plugin",{"_creationTime":210,"_id":211,"community":212,"display":213,"identity":216,"providers":218,"relations":227,"workflow":229},1778053782268.1958,"k1787v8f6tx7dxy8c4xfz7hk2h867f5j",{"reviewCount":8},{"description":198,"installMethods":214,"name":200,"sourceUrl":201,"tags":215},{},[20,21,15,203,204],{"githubOwner":188,"githubRepo":189,"locale":25,"slug":189,"type":217},"marketplace",{"extract":219,"llm":224},{"commitSha":220,"license":221,"marketplace":222},"648df5daa32ef7a742a07740d9c5d13e3b8229c4","MIT",{"name":189,"pluginCount":223},1,{"promptVersionExtension":168,"promptVersionScoring":169,"score":225,"targetMarket":174,"tier":226},70,"evaluated",{"repoId":228},"kd74nzm561ws7px2qx08z27fad864yft",{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},1778053782949,1778053782268,{"extract":233,"llm":234},{"commitSha":220,"license":221},{"promptVersionExtension":168,"promptVersionScoring":169,"score":235,"targetMarket":174,"tier":226},75,{"parentExtensionId":211,"repoId":228},{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},{"extract":239,"llm":240},{"commitSha":220,"license":221},{"promptVersionExtension":168,"promptVersionScoring":169,"score":171,"targetMarket":174,"tier":175},{"parentExtensionId":195,"repoId":228},{"_creationTime":243,"_id":228,"identity":244,"providers":245,"workflow":256},1777995558409.8423,{"githubOwner":188,"githubRepo":189,"sourceUrl":201},{"discover":246,"github":249},{"sources":247},[248],"skills-sh",{"closedIssues90d":250,"forks":251,"openIssues90d":8,"pushedAt":252,"readmeSize":253,"stars":254,"topics":255},2,58,1777486877000,3660,651,[],{"discoverAt":257,"extractAt":258,"githubAt":258,"updatedAt":258},1777995558409,1778053783581,{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},[],[262,280,301,323,356,386],{"_creationTime":263,"_id":264,"community":265,"display":266,"identity":274,"providers":275,"relations":278,"workflow":279},1778053782268.1982,"k17c7a6x1535rf64abc4aj73d18673he",{"reviewCount":8},{"description":267,"installMethods":268,"name":269,"sourceUrl":270,"tags":271},"INVOKE THIS SKILL at the START of any LangChain/LangGraph/Deep Agents project, before writing any agent code. Determines which framework layer is right for the task: LangChain, LangGraph, Deep Agents, or a combination. Must be consulted before other agent skills.",{},"Framework Selection","https://github.com/langchain-ai/langchain-skills/tree/HEAD/config/skills/framework-selection",[20,21,15,272,273],"framework-selection","documentation",{"githubOwner":188,"githubRepo":189,"locale":25,"slug":272,"type":191},{"extract":276,"llm":277},{"commitSha":220,"license":221},{"promptVersionExtension":168,"promptVersionScoring":169,"score":171,"targetMarket":174,"tier":175},{"parentExtensionId":195,"repoId":228},{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},{"_creationTime":281,"_id":282,"community":283,"display":284,"identity":293,"providers":295,"relations":299,"workflow":300},1778053782268.1968,"k17b4azxqtakm78mxatx1sst0d867h79",{"reviewCount":8},{"description":285,"installMethods":286,"name":287,"sourceUrl":288,"tags":289},"INVOKE THIS SKILL when building ANY Deep Agents application. Covers create_deep_agent(), harness architecture, SKILL.md format, and configuration options.",{},"Deep Agents Core","https://github.com/langchain-ai/langchain-skills/tree/HEAD/config/skills/deep-agents-core",[15,20,21,290,291,292],"agent-framework","python","typescript",{"githubOwner":188,"githubRepo":189,"locale":25,"slug":294,"type":191},"deep-agents-core",{"extract":296,"llm":297},{"commitSha":220,"license":68},{"promptVersionExtension":168,"promptVersionScoring":169,"score":298,"targetMarket":174,"tier":175},93,{"parentExtensionId":195,"repoId":228},{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},{"_creationTime":302,"_id":303,"community":304,"display":305,"identity":315,"providers":317,"relations":321,"workflow":322},1778053782268.2012,"k1758g6mkayw5kxre5409yjfkd866gmj",{"reviewCount":8},{"description":306,"installMethods":307,"name":308,"sourceUrl":309,"tags":310},"INVOKE THIS SKILL when implementing human-in-the-loop patterns, pausing for approval, or handling errors in LangGraph. Covers interrupt(), Command(resume=...), approval/validation workflows, and the 4-tier error handling strategy.",{},"LangGraph Human-in-the-Loop","https://github.com/langchain-ai/langchain-skills/tree/HEAD/config/skills/langgraph-human-in-the-loop",[21,19,311,312,313,314],"workflow","interrupt","approval","error-handling",{"githubOwner":188,"githubRepo":189,"locale":25,"slug":316,"type":191},"langgraph-human-in-the-loop",{"extract":318,"llm":319},{"commitSha":220,"license":221},{"promptVersionExtension":168,"promptVersionScoring":169,"score":320,"targetMarket":174,"tier":175},90,{"parentExtensionId":195,"repoId":228},{"anyEnrichmentAt":230,"extractAt":231,"githubAt":230,"llmAt":186,"updatedAt":186},{"_creationTime":324,"_id":325,"community":326,"display":327,"identity":337,"providers":340,"relations":349,"workflow":351},1778053968286.4963,"k170epaxhy76k502zg91c998t1867cwh",{"reviewCount":8},{"description":328,"installMethods":329,"name":330,"sourceUrl":331,"tags":332},"Searches multiple web sources, synthesizes findings, and produces cited research reports using delegated subagents. Use when the user asks to research a topic online, search the web, look something up, find current information, compare options, or produce a research report.",{},"Web Research Skill","https://github.com/langchain-ai/deepagents/tree/HEAD/libs/cli/examples/skills/web-research",[333,334,335,336,17,273,20],"research","productivity","web-search","synthesis",{"githubOwner":188,"githubRepo":338,"locale":25,"slug":339,"type":191},"deepagents","web-research",{"extract":341,"llm":344,"smithery":346},{"commitSha":342,"license":343},"b108c71d0c570e16c7050c1eac482e15dc35a5ed","MIT-0",{"promptVersionExtension":168,"promptVersionScoring":169,"score":345,"targetMarket":174,"tier":175},85,{"qualityScore":347,"totalActivations":348,"uniqueUsers":348,"useCount":8,"verified":183},0.7393806,8,{"repoId":350},"kd76dna2fvfbnjvzcpd2cwqnyd865xz7",{"anyEnrichmentAt":352,"extractAt":353,"githubAt":354,"llmAt":355,"smitheryAt":352,"updatedAt":355},1778053994907,1778053968286,1778053969344,1778054053159,{"_creationTime":357,"_id":358,"community":359,"display":360,"identity":371,"providers":375,"relations":380,"workflow":382},1778053622473.6682,"k171xwg814163s51767b0q0pn1866cwe",{"reviewCount":8},{"description":361,"installMethods":362,"name":363,"sourceUrl":364,"tags":365},"Orchestrate multiple worker agents to implement groomed tasks in Gitea repositories. Use when multiple ready tasks need implementation, when you want autonomous multi-task execution, or when coordinating batch development work with Gitea. Keywords: coordinator, orchestrator, multi-task, parallel, workers, batch, autonomous, gitea, tea.",{},"Gitea Coordinator","https://github.com/jwynia/agent-skills/tree/HEAD/skills/tech/development/workflow/gitea-coordinator",[366,16,367,368,369,311,370],"gitea","automation","agile","development","multi-agent",{"githubOwner":372,"githubRepo":373,"locale":25,"slug":374,"type":191},"jwynia","agent-skills","gitea-coordinator",{"extract":376,"llm":378},{"commitSha":377,"license":221},"e02ec7e226a6e4f8419fd3b88a1d8e472d421b32",{"promptVersionExtension":168,"promptVersionScoring":169,"score":379,"targetMarket":174,"tier":175},98,{"repoId":381},"kd7efn3mprpa8rd8vm5hw5ebzx864fph",{"anyEnrichmentAt":383,"extractAt":384,"githubAt":383,"llmAt":385,"updatedAt":385},1778053625386,1778053622473,1778054012696,{"_creationTime":387,"_id":388,"community":389,"display":390,"identity":398,"providers":400,"relations":403,"workflow":404},1778053622473.6672,"k178b0dafhyecx3y6cbxkne8mh866zc4",{"reviewCount":8},{"description":391,"installMethods":392,"name":393,"sourceUrl":394,"tags":395},"Orchestrate multiple worker agents to implement groomed tasks. Use when multiple ready tasks need implementation, when you want autonomous multi-task execution, or when coordinating batch development work. Keywords: coordinator, orchestrator, multi-task, parallel, workers, batch, autonomous.",{},"Agile Coordinator","https://github.com/jwynia/agent-skills/tree/HEAD/skills/tech/development/workflow/agile-coordinator",[368,311,16,396,397,367],"developer-tools","git",{"githubOwner":372,"githubRepo":373,"locale":25,"slug":399,"type":191},"agile-coordinator",{"extract":401,"llm":402},{"commitSha":377,"license":221},{"promptVersionExtension":168,"promptVersionScoring":169,"score":379,"targetMarket":174,"tier":175},{"repoId":381},{"anyEnrichmentAt":383,"extractAt":384,"githubAt":383,"llmAt":385,"updatedAt":385}]