langchain

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/llm-tools/langchain/SKILL.md

Quality

Content

57%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured with good progressive disclosure to real reference files, but it is over-long for a SKILL.md and teaches a mix of deprecated LangChain APIs alongside the current create_agent() API, reducing actionability and conciseness.

Suggestions

Remove deprecated APIs (LLMChain, ConversationChain, RetrievalQA, ConversationBufferMemory, create_tool_calling_agent/AgentExecutor, load_qa_with_sources_chain, chain.run(), dict-style chain calls) and standardize on the current langchain create_agent() / LCEL patterns, or move legacy snippets into a clearly labeled 'deprecated/old patterns' section.

Cut the metrics block (GitHub stars, contributor counts), the performance-benchmarks table, and the LangChain-vs-LangGraph comparison from SKILL.md — these are marketing/context that don't earn their tokens, and the LangGraph guidance already lives under 'Use alternatives instead'.

Trim the in-body Agents and RAG sections to quick-start essentials and delegate depth to references/agents.md and references/rag.md (which already cover the same ground), keeping SKILL.md an overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient executable snippets, but padded with a metrics block ("119,000+ GitHub stars", "3,800+ contributors"), a performance-benchmarks table, and a LangChain-vs-LangGraph comparison, plus full agent/RAG sections that duplicate the references — could be tightened.	2 / 3
Actionability	Code is concrete and copy-pasteable, but leans heavily on deprecated APIs — LLMChain, ConversationChain, RetrievalQA, ConversationBufferMemory, create_tool_calling_agent/AgentExecutor, load_qa_with_sources_chain, chain.run(), and dict-style qa_chain({...}) — alongside an eval()-based calculator tool, so the guidance is partly misleading rather than fully current.	2 / 3
Workflow Clarity	The RAG pipeline is laid out as a numbered sequence (load, split, embed, retriever, QA, query), but there are no validation checkpoints or error-recovery feedback loops, and nothing confirms the index/outputs before use.	2 / 3
Progressive Disclosure	A clear References section points to real, one-level-deep files (references/agents.md, references/rag.md, references/integration.md) that exist on disk and are well-signaled by topic, giving easy navigation.	3 / 3
	Total	9 / 12 Passed

Description

92%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, concrete description that answers both what LangChain does and when to use it with natural trigger terms. Its only weak spot is distinctiveness, since the agent/RAG framing overlaps with sibling LLM-application frameworks.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "building LLM-powered applications with agents, chains, and RAG", "ReAct agents, tool calling, memory management, and vector store retrieval" — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly answers both what ("Framework for building LLM-powered applications...") and when ("Use for building chatbots, question-answering systems, autonomous agents, or RAG applications"), matching the anchor with an explicit trigger clause.	3 / 3
Trigger Term Quality	Covers natural terms a user would say — "chatbots", "question-answering systems", "autonomous agents", "RAG applications", "LLM-powered applications" — with good variation across the space.	3 / 3
Distinctiveness Conflict Risk	The skill is LangChain-named with a defined niche, but the description's coverage of agents and RAG overlaps materially with named alternatives (LlamaIndex, LangGraph) noted in the body, so it could still trigger for a sibling skill.	2 / 3
	Total	11 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/langchain/SKILL.md
Commit: 52845c3

Reviewed: about 14 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.