langchain-architecture

Design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration. Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows.

1.66x

Quality

68%

Does it follow best practices?

Impact

83%

1.66x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/langchain-architecture/SKILL.md

The canonical home for this skill is langchain-architecture in wshobson/agents

Quality

Content

57%Weight 40%Scale 1-5

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

The content is highly actionable with abundant near-executable code covering the major LangChain/LangGraph patterns, but it is a monolithic reference document: concept-list padding, no bundle-file offloading, and no sequenced workflow with integrated validation checkpoints hold it back.

Suggestions

Move the architecture patterns, memory management, testing, and performance sections into separate files under references/ (e.g., references/patterns.md, references/memory.md, references/testing.md) and keep SKILL.md as a concise overview with clearly signaled one-level-deep links.

Trim conceptual listings Claude already knows (memory class taxonomy, document-processing components, LangSmith feature bullets) and drop time-sensitive version/date claims like 'langchain (1.2.x)' and 'the standard for building agents in 2026' or confine them to a versioned 'deprecated/old patterns' section.

Make every code example self-contained by defining the free variables (llm, research_tools, writing_tools, text_splitter, embeddings_model) or explicitly marking them as placeholders, so snippets are copy-paste ready.

Dimension	Reasoning	Score
Conciseness	The bulk is concrete code, but padded concept listings Claude already knows ('LangGraph is the standard for building agents in 2026', the ConversationBufferMemory-style memory class list, the Document Loaders/Text Splitters component list) and time-sensitive version info ('langchain (1.2.x)') outside any deprecated section keep it at 'mostly efficient but includes some unnecessary explanation'.	3 / 5
Actionability	Nearly all examples are executable — a full create_react_agent setup with a safe AST-based calculator, a complete RAG StateGraph, StructuredTool with Pydantic schemas, streaming, and pytest tests — but several snippets reference undefined variables (llm, research_tools, text_splitter, embeddings_model), matching 'concrete code with minor gaps' rather than fully copy-paste ready.	4 / 5
Workflow Clarity	The body is a topic catalog of independent patterns rather than a sequenced procedure; validation material exists (Testing Strategies, Production Checklist) but is not integrated as checkpoints inside any workflow, matching 'sequence present but checkpoints missing or implicit'.	3 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent), so ~400 lines of patterns, memory, testing, and performance content that belongs in separate reference files is inlined in one 660-line SKILL.md; section headers give it structure, matching 'content that should be separate is inline' with some organization.	3 / 5
	Total	13 / 20 Passed

Description

78%Weight 40%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.

A strong description that explicitly states both what the skill does and when to use it, with concrete, natural trigger phrases. Its only weakness is that it names one action verb plus capability areas rather than several distinct concrete actions.

Dimension	Reasoning	Score
Specificity	Only one concrete verb ('Design LLM applications') is given; 'agents, memory, and tool integration' are capability nouns rather than distinct actions, matching the anchor 'names domain and 1-2 concrete actions' rather than the several-action anchor above.	3 / 5
Completeness	It explicitly answers both questions: what ('Design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration') and when ('Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows') with three concrete trigger phrases, mirroring the 5-anchor example's structure.	5 / 5
Trigger Term Quality	Natural terms users would say ('LangChain', 'LangGraph', 'AI agents', 'LLM workflows') are all present, matching 'good keyword coverage; a few natural terms missing' — synonyms like 'RAG', 'chatbot', or 'chains' are absent, so it falls short of the 5 anchor.	4 / 5
Distinctiveness Conflict Risk	Naming 'LangChain 1.x and LangGraph' carves a clear framework niche, but the broader 'LLM applications / agents / workflows' framing overlaps other LLM-architecture skills, matching 'mostly distinct; minor overlap risk' rather than the minimal-conflict 5 anchor.	4 / 5
	Total	16 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (667 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: Dicklesworthstone/pi_agent_rust
Path: tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/langchain-architecture/SKILL.md
Commit: a65ce00

Reviewed: 6 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.