langchain-architecture

Master the LangChain framework for building sophisticated LLM applications with agents, chains, memory, and tool integration.

Quality

40%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/langchain-architecture/SKILL.md

Quality

Content

43%Scale 1-5

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

This skill provides a broad survey of LangChain features with concrete code examples, but suffers from significant verbosity by cataloging concepts Claude already knows (agent types, memory types, chain types). The lack of coherent multi-step workflows with validation checkpoints is a notable weakness for a skill that covers document processing pipelines and agent execution. The referenced bundle files don't exist, undermining the progressive disclosure structure.

Suggestions

Remove or drastically condense the 'Core Concepts' taxonomy section — Claude already knows what LangChain agents, chains, and memory types are. Keep only non-obvious guidance or gotchas.

Add explicit multi-step workflows with validation checkpoints, especially for document processing pipelines (e.g., load → split → embed → validate embedding count → store → test retrieval).

Update code examples to use current LangChain APIs (e.g., `langchain_openai.ChatOpenAI` instead of deprecated `langchain.llms.OpenAI`, and LCEL instead of `initialize_agent`).

Provide the referenced bundle files (references/agents.md, assets/agent-template.py, etc.) and move the inlined reference material into them to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The skill is significantly verbose, listing out agent types, chain types, memory types, and document processing components that Claude already knows well. The 'Core Concepts' section is essentially a taxonomy of LangChain features that adds little actionable value. The entire file is ~250 lines when it could be much leaner by deferring reference material to the referenced files.	2 / 5
Actionability	The code examples are concrete and mostly executable — the Quick Start, RAG pattern, custom agent, sequential chain, memory selection, callbacks, testing, and performance sections all contain real Python code. However, some examples use deprecated APIs (e.g., `initialize_agent`, `from langchain.llms import OpenAI`) and tool implementations have placeholder comments rather than real logic, which slightly reduces copy-paste readiness.	4 / 5
Workflow Clarity	There is no clear multi-step workflow with validation checkpoints. The skill presents isolated code snippets and patterns but never sequences them into a coherent workflow (e.g., 'first do X, validate, then do Y'). The batch processing section lacks validation/verification steps, and the production checklist is a static list rather than an integrated workflow. For a skill involving agent execution and document processing pipelines, the absence of error recovery loops and validation steps is a significant gap.	2 / 5
Progressive Disclosure	The skill references multiple external files (references/agents.md, assets/agent-template.py, etc.) and mentions resources/implementation-playbook.md, which is good structure. However, no bundle files are provided, so these references are unverifiable. Additionally, the Core Concepts section inlines substantial reference material (agent types, chain types, memory types) that should be in the referenced files, making the main file bloated.	3 / 5
	Total	11 / 20 Passed

Description

36%Scale 1-5

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

The description identifies the domain (LangChain) and lists high-level concepts but lacks concrete actions, explicit trigger guidance, and a 'Use when' clause. The imperative 'Master' framing is inappropriate for a skill description that should tell Claude what the skill does and when to use it. It reads more like a course title than a functional skill description.

Suggestions

Add an explicit 'Use when...' clause with trigger phrases like 'Use when the user asks about LangChain, building LLM pipelines, RAG applications, or agent workflows'.

Replace 'Master the LangChain framework' with concrete actions in third person, e.g., 'Creates LangChain agents, builds retrieval-augmented generation pipelines, configures prompt templates and memory, and integrates external tools'.

Include additional natural trigger terms and synonyms such as 'RAG', 'retrieval', 'prompt templates', 'vector stores', 'LCEL', and 'langchain' to improve keyword coverage.

Dimension	Reasoning	Score
Specificity	Names the domain (LangChain framework) and lists high-level concepts (agents, chains, memory, tool integration), but these are abstract categories rather than concrete actions. No specific actions like 'create chains', 'configure agents', or 'set up retrieval pipelines' are described.	2 / 5
Completeness	Has a vague 'what' (building LLM applications with LangChain) but completely lacks any 'when' clause or trigger guidance. The word 'Master' frames it as a learning goal rather than a functional description, and there is no 'Use when...' or equivalent.	2 / 5
Trigger Term Quality	Includes 'LangChain', 'agents', 'chains', 'memory', 'tool integration', and 'LLM applications' which are relevant keywords. However, it misses natural user phrases like 'RAG', 'retrieval', 'prompt templates', 'vector stores', 'langchain.js', or 'LCEL'.	3 / 5
Distinctiveness Conflict Risk	LangChain is a specific framework which provides some distinctiveness, but terms like 'LLM applications', 'agents', and 'tool integration' are broad enough to overlap with general AI/LLM development skills or other framework-specific skills.	3 / 5
	Total	10 / 20 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

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

	Total	10 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: skills/langchain-architecture/SKILL.md
Commit: 57c135b

Reviewed: 4 days 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.