LangGraph Workflow Development Skill

SkillDev tools

Guide for LangGraph workflows in AI-powered test automation

Use LangGraph Workflow Development Skill in Claude, ChatGPT or Ahel Desktop

Free. Sign in, add LangGraph Workflow Development Skill and connect your AI. About a minute.

Also: Claude Code · Cursor · Codex

Then ask your AI: use the LangGraph Workflow Development Skill

Details

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

LangGraph Workflow Development SkillStart free

What this skill tells your AI

The instructions your AI receives, as published by aiqualitylab/ai-natural-language-tests in .github/skills/langgraph-workflow-development/SKILL.md and read by Ahel’s review.

Workflow Pipeline

Step2FetchData -> Step3SearchPatterns -> Step4GenerateTests -> Step5RunTests(optional) -> END

State Schema

@dataclass
class TestState:
    requirements: List[str]
    output_dir: str
    use_prompt: bool
    approve: bool = False
    framework: str = "cypress"
    url: Optional[str] = None
    run_tests: bool
    llm_provider: str = "openai"
    test_data: Optional[Dict] = None
    context: str = ""
    similar_patterns: List = field(default_factory=list)
    generated_tests: List = field(default_factory=list)
    test_results: Optional[Dict] = None
    run_id: str = ""

Building Workflow

from langgraph.graph import StateGraph, END

def create_workflow() -> StateGraph:
    workflow = StateGraph(TestState)

    workflow.add_node("step_2", step_2_fetch_test_data)
    workflow.add_node("step_3", step_3_search_similar_patterns)
    workflow.add_node("step_4", step_4_generate_tests)
    workflow.add_node("step_5", step_5_run_tests)

    workflow.set_entry_point("step_2")
    workflow.add_edge("step_2", "step_3")
    workflow.add_edge("step_3", "step_4")
    workflow.add_conditional_edges("step_4", should_run_tests, {"run_tests": "step_5", END: END})
    workflow.add_edge("step_5", END)

    return workflow.compile(checkpointer=WORKFLOW_CHECKPOINTER)

Node Pattern

def step_4_generate_tests(state: TestState) -> TestState:
    # Current implementation uses RunnableLambda for prompt generation
    # and BaseOutputParser for code-fence extraction before file save.
    return state

Mode Selection

# Prompt mode applies only when framework supports it (currently Cypress)
use_prompt_mode = state.use_prompt and FRAMEWORK_CONFIG[state.framework]["supports_prompt_mode"]

Context Sources

# URL analysis in step_2 builds test_data and context block
# Similar pattern retrieval in step_3 augments context
# Final context is consumed by step_4 prompt template rendering

Error Handling

def node_with_error_handling(state):
    try:
        result = process(state)
        return result
    except Exception as e:
        state.error = str(e)
        return state

Signals

GitHub stars
23
Forks
7
Last commit
Oct 2026
Advanced
Item type
skill
Key
langgraph-workflow-development
Source
github.com/aiqualitylab/ai-natural-language-tests