deepagents
LangChain Deep Agents framework for building autonomous coding agents. Use for agent harness, backends, subagents, human-in-the-loop, long-term memory, middleware, and CLI-based agent development.
npx skills add majiayu000/claude-skill-registry --skill deepagents --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
This skill describes how to build autonomous coding agents using LangChain Deep Agents. It explains how to configure agent backends (StateBackend, FilesystemBackend, StoreBackend, CompositeBackend) and how to route file operations across backends. It outlines supported tooling (ls, read_file, write_file, edit_file, glob, grep) and how to wire a backend into create_deep_agent, including virtual filesystem concepts and policy hooks.
How it works
The skill details concrete backends and their behavior:
- StateBackend: Ephemeral storage, accessed via runtime state; persists across agent turns on the same thread through checkpoints.
- FilesystemBackend (local disk): Reads/writes real files under a root_dir; optional virtual_mode for sandboxed paths; uses secure path resolution.
- StoreBackend (LangGraph store): Uses a LangGraph BaseStore for cross-thread durable storage; suitable for long-term memories.
- CompositeBackend (router): Routes paths to different backends (e.g., "/memories/" to StoreBackend, others to StateBackend); preserves original path prefixes; supports multiple backends for a single filesystem namespace.
- Route to backends: You can pass either a BackendProtocol instance or a BackendFactory to create_deep_agent(backend=...). If omitted, defaults to StateBackend(rt).
- Policy hooks: Can extend backends with Guards or PolicyWrapper to enforce rules on writes/edits.
- Protocol and operations: LS, read, grep, glob, write, and edit endpoints are defined; examples show path handling, error strings, and how results should be structured.
When to use it
Trigger this skill when working with deepagents, asking about deepagents features or APIs, implementing deepagents solutions, debugging deepagents code, or learning deepagents best practices.
What it can touch
The skill references specific backends and tools: StateBackend, FilesystemBackend, StoreBackend, CompositeBackend, create_deep_agent, and tools like ls, read_file, write_file, edit_file, glob, grep. It notes how backends are configured and how routes map to different backends. It also shows how to pass a store to create_deep_agent and how to access runtime components like runtime.state and runtime.store.
Caveats
The description includes guidance on implementing custom backends (S3, Postgres) and examples of policy wrappers. It emphasizes that different backends may return files_update semantics differently (state backends may return files_update; external backends may return None). It also notes that longer prefixes win for routing decisions. The content is instruction-based and expects implementers to follow the specified endpoints and behaviors; no guarantees of outcomes are provided.
# Deepagents Skill Langchain deep agents framework for building autonomous coding agents. use for agent harness, backends, subagents, human-in-the-loop, long-term memory, middleware, and cli-based agent development., generated from official documentation. ## When to Use This Skill This skill should be triggered when: - Working with deepagents - Asking about deepagents features or APIs - Implementing deepagents solutions - Debugging deepagents code - Learning deepagents best practices ## Quick Reference ### Common Patterns **Pattern 1:** Docs by LangChain home pageLangChain + LangGraphSearch...⌘KSupportGitHubTry LangSmithTry LangSmithSearch...NavigationCore capabilitiesBackendsLangChainLangGraphDeep AgentsIntegrationsLearnReferenceContributePythonOverviewGet startedQuickstartCustomizationCore capabilitiesAgent harnessBackendsSubagentsHuman-in-the-loopLong-term memoryMiddlewareCommand line interfaceUse the CLIOn this pageQuickstartBuilt-in backendsStateBackend (ephemeral)FilesystemBackend (local disk)StoreBackend (LangGraph store)CompositeBackend (router)Specify a backendRoute to different backendsUse a virtual filesystemAdd policy hooksProtocol referenceCore capabilitiesBackendsCopy
- When to Use This Skill
- Quick Reference
- Common Patterns
- Example Code Patterns
- Reference Files
- Working with This Skill
- For Beginners
- For Specific Features
- For Code Examples
- Resources
- references/
- scripts/
- assets/
- Notes
What does the deepagents skill do?
LangChain Deep Agents framework for building autonomous coding agents. Use for agent harness, backends, subagents, human-in-the-loop, long-term memory, middleware, and CLI-based agent development.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill deepagents --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.