@langchain-ai

langgraph-persistence covers checkpointer setup (in-memory, sqlite, postgres), thread isolation, state history replay, subgraph scoping rules, and cross-thread long-term memory via store. includes common pitfalls and reducer semantics.

langgraph-fundamentals covers stategraph design, node and edge patterns, state management with reducers, command-driven routing, send-based fan-out, streaming modes, and retry/error handling strategies across python and typescript.

langchain-middleware provides human-in-the-loop approval for agent tool calls with per-tool policies, decision resumption (approve/edit/reject), and custom middleware hooks for retry logic and guardrails. requires checkpointer and thread_id config.

langchain-rag covers the full retrieval-augmented generation pipeline from document loading through embedding, storage, and retrieval. provides working code samples in python and javascript with common pitfalls highlighted.

web-research delegates multi-subtopic investigations to parallel subagents, aggregates findings from local files, and synthesizes cited reports. execution relies on explicit planning and file-based communication between agents.

blog-post guides agents through researching topics, structuring markdown content, generating cover images, and optimizing for seo. covers workflow delegation, required output artifacts, and content templates.

langchain-ai-web-research delegates multi-subtopic web research to parallel subagents, collects findings into local files, and synthesizes cited reports. covers planning, delegation, and synthesis workflows with explicit file-based communication patterns.

schema-exploration maps database structure by listing tables, describing columns and data types, identifying foreign key relationships, and documenting entity relationships. oriented toward understanding schema before querying.

deep-agents-memory provides pluggable backends for agent state and file operations, supporting ephemeral thread-scoped storage, cross-session persistence, and hybrid routing via longest-prefix matching.

langchain-dependencies guides selection between langchain versions, core foundation setup, orchestration framework choice, and isolated provider integration installs with semantic versioning caveats.

langsmith-dataset covers creation, management, and upload of evaluation datasets across four types (final_response, single_step, trajectory, rag) via cli and sdk, with auth via api key and project configuration.

deep-agents-core documents the harness architecture, create_deep_agent() api, middleware configuration (todo tracking, filesystem, subagents, memory, approval workflows), skill loading, and common configuration errors across python and javascript.

framework-selection provides a three-tier mental model for choosing between langchain, langgraph, and deep agents based on agent complexity and feature needs, but lacks concrete decision criteria and rollback guidance.

langgraph-docs fetches documentation indexes and surfaces relevant pages for building stateful agents and multi-agent workflows, but lacks decision logic and recovery patterns for common fetch failures.

langchain-fundamentals covers agent creation via create_agent with tool binding, state checkpointing, and middleware integration for human oversight and error recovery patterns.

ecosystem-primer maps the langchain/langgraph/deep agents technology stack and guides selection between abstraction layers based on execution durability and control flow needs.

social-media drafts platform-specific posts for linkedin and twitter with templated hooks and auto-generated images. requires explicit research delegation before writing. platform-specific character limits enforced.

skill-creator documents where the deepagents CLI loads skills from and how precedence works between user and project directories. lacks guidance on actual skill authoring.

langsmith-evaluator exposes environment variable setup and a single CLI command for listing evaluators, but omits intent, procedure steps, decision logic, and output contracts entirely.

INVOKE THIS SKILL when working with LangSmith tracing OR querying traces. Covers adding tracing to applications and querying/exporting trace data. Uses the…

Dispatches many independent items in parallel: create a table, fan out to subagents, aggregate results. One row = one unit of work.

INVOKE THIS SKILL when building, testing, or deploying Managed Deep Agents in LangSmith with the mda CLI. Walks a user through their first agent end to end —…

INVOKE THIS SKILL when using the langgraph CLI to scaffold, develop, build, or deploy LangGraph applications. Covers langgraph new, dev, build, up, deploy, and…

INVOKE THIS SKILL when using subagents, task planning, or human approval in Deep Agents. Covers SubAgentMiddleware, TodoList for planning, and HITL interrupts.

INVOKE THIS SKILL when implementing human-in-the-loop patterns, pausing for approval, or handling errors in LangGraph. Covers interrupt(), Command(resume=...),…

Guide for creating effective skills that extend agent capabilities with specialized knowledge, workflows, or tool integrations. Use this skill when the user asks to: (1) create a new skill, (2) make a skill, (3) build a skill, (4) set up a skill, (5) initialize a skill, (6) scaffold a skill, (7) update or modify an existing skill, (8) validate a skill, (9) learn about skill structure, (10) understand how skills work, or (11) get guidance on skill design patterns. Trigger on phrases like "create a skill", "new skill", "make a skill", "skill for X", "how do I create a skill", or "help me build a skill".

Writes and structures long-form blog posts, creates tutorial outlines, and optimizes content for SEO with cover image generation. Use when the user asks to write a blog post, article, how-to guide, tutorial, technical writeup, thought leadership piece, or long-form content.

Search arXiv preprint repository for papers in physics, mathematics, computer science, quantitative biology, and related fields

Drafts engaging social media posts, writes hooks, suggests hashtags, creates thread structures, and generates companion images. Use when the user asks to write a LinkedIn post, tweet, Twitter/X thread, social media caption, social post, or repurpose content for social platforms.

Searches arXiv for preprints and academic papers, retrieves abstracts, and filters by topic. Use when the user asks to find research papers, search arXiv, look…

Writes and executes SQL queries from simple SELECTs to complex multi-table JOINs, aggregations, and subqueries. Use when the user asks to query a database,…