fsd:debug-flow

Diagnose AI flow failures by analyzing NDJSON traces and stderr logs.

2|Updated Feb 16, 2026
One-click install
npx skills add https://github.com/fixpoint-labs/flow-state-dev --skill fsd-debug-flow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fsd:debug-flow
Source: https://github.com/fixpoint-labs/flow-state-dev/tree/main/.agents/skills/debug-flow
Command: npx skills add https://github.com/fixpoint-labs/flow-state-dev --skill fsd-debug-flow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @flow-state-dev/cli, and includes scripts (resource) components.

What problem does it solve?

This Skill streamlines the debugging process for AI flows, allowing users to quickly identify and resolve issues by analyzing execution traces and logs.

Core Features & Use Cases

  • Flow Execution Tracing: Analyze NDJSON traces and stderr logs to isolate failing blocks and diagnose root causes.
  • CLI Execution: Run flows via the CLI to capture full execution outputs and analyze them in detail.
  • Isolation Testing: Execute blocks in isolation to test suspected failing blocks and validate fixes.
  • Use Case: If a chatbot in your application is not functioning correctly, use this Skill to trace the execution, identify the failing block, and diagnose the issue.

Quick Start

Run the 'fsd:debug-flow' Skill with the specific flow and action details, providing a description of the problem to analyze.

Frequently Asked Questions about fsd:debug-flow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug an AI flow execution failure?

To debug an AI flow execution failure, you can run the flow through the CLI to capture NDJSON traces and stderr logs. This process isolates failing blocks and diagnoses root causes like schema mismatches or model errors.

What causes schema mismatches and model errors during flow execution?

Schema mismatches and model errors during flow execution are common issues that occur within processing blocks. Analyzing stderr logs and NDJSON execution traces helps isolate the specific failing block to diagnose these root causes.

How do I test a single failing block in isolation?

You can test a suspected failing block in isolation by executing it independently via the @flow-state-dev CLI. This isolation testing captures specific execution outputs to validate fixes without running the entire flow.

Do I need the @flow-state-dev framework to trace execution issues?

Yes, using the @flow-state-dev framework and CLI is required to trace execution issues. The diagnostic process depends on the CLI to run flows, capture NDJSON traces, and analyze stderr logs for root cause isolation.

Why is my chatbot flow timing out and how can I trace it?

Chatbot flow timeouts occur during execution and can be traced by running the flow via the CLI. Analyzing the generated NDJSON traces and stderr logs isolates the failing block causing the timeout to diagnose the issue.