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.
74
66%
Does it follow best practices?
Impact
83%
1.66xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/langchain-architecture/SKILL.mdQuality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid skill description that clearly identifies its technology niche (LangChain 1.x / LangGraph) and provides explicit trigger guidance. Its main weakness is that the 'what' portion could be more specific about concrete actions beyond the high-level 'Design' verb. The trigger terms are well-chosen and the description is concise without being vague.
Suggestions
Replace the vague 'Design' with more specific concrete actions, e.g., 'Build chains, configure retrieval pipelines, set up agent tool-calling, and manage conversational memory using LangChain 1.x and LangGraph.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (LLM applications, LangChain, LangGraph) and mentions some capabilities (agents, memory, tool integration), but doesn't list multiple concrete actions—'Design' is somewhat vague and there are no specific operations like 'create chains', 'configure retrievers', or 'set up streaming'. | 2 / 3 |
Completeness | Clearly answers both 'what' (design LLM applications using LangChain 1.x and LangGraph for agents, memory, and tool integration) and 'when' (explicit 'Use when building LangChain applications, implementing AI agents, or creating complex LLM workflows'). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'LangChain', 'LangGraph', 'AI agents', 'LLM workflows', 'memory', 'tool integration'. These cover the main terms a developer would use when seeking help with this technology stack. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to specific technology references (LangChain 1.x, LangGraph) which create a clear niche. Unlikely to conflict with generic coding skills or other LLM framework skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, executable code examples across multiple LangChain/LangGraph patterns, which is its primary strength. However, it is severely bloated—functioning more as comprehensive documentation than a focused skill file—explaining many concepts Claude already understands and inlining content that should be split across bundle files. It also lacks validation checkpoints and error recovery workflows for the multi-step processes it describes.
Suggestions
Reduce content by 60-70%: remove the 'Core Concepts' descriptive section, the 'When to Use This Skill' list, and explanatory text around code blocks. Keep only the Quick Start agent and 1-2 key patterns inline.
Split detailed patterns (RAG, Multi-Agent, Multi-Step Workflow), memory management, callbacks, streaming, and testing into separate bundle files (e.g., PATTERNS.md, MEMORY.md, TESTING.md) with clear one-level references from SKILL.md.
Add explicit validation/verification steps to workflows—e.g., after building a StateGraph, compile and test with a sample input, check for expected state transitions, and handle compilation errors before proceeding.
Remove bullet-point feature lists (Memory Systems, Document Processing, Callbacks & Tracing) that describe what things are rather than providing actionable instructions Claude doesn't already know.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~500+ lines. Explains concepts Claude already knows (what StateGraph is, what memory systems do, what callbacks are). Lists features like bullet-point documentation rather than providing targeted, novel guidance. The 'Core Concepts' section is largely descriptive padding. Multiple patterns could be consolidated or referenced externally. | 1 / 3 |
Actionability | The code examples are concrete, executable, and copy-paste ready. Tools are defined with full implementations (e.g., the safe math evaluator), state graphs are fully wired with START/END edges, and patterns include complete working examples with imports and async invocation. | 3 / 3 |
Workflow Clarity | Multi-step patterns (RAG, multi-agent, multi-step workflow) show clear sequencing through StateGraph edges, but there are no explicit validation checkpoints, error recovery loops, or verification steps. The production checklist is a static list rather than a sequenced workflow with feedback loops. | 2 / 3 |
Progressive Disclosure | Monolithic wall of text with no bundle files to offload content. All patterns, memory management, callbacks, streaming, testing, and performance optimization are inlined in a single massive file. No references to supporting files for detailed patterns, and the external links are just documentation URLs rather than structured bundle references. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%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 |
|---|---|---|
skill_md_line_count | SKILL.md is long (667 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
bbc5ade
- Reviewed
Table of Contents
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.