---
title: Langfuse for workflow automation
description: Trace multi-step workflow automation across systems and frameworks. Track cost at every step and include human expertise to improve your agents over time.
contentWidth: full
---

<h1 className="sr-only">Langfuse for workflow automation</h1>

  <section data-use-case-section="hero" className="grid border-b border-line-structure lg:grid-cols-[1.2fr_0.8fr]">
    
      <p className="font-mono text-[10px] uppercase tracking-[0.09em] text-text-tertiary">
        Use case / workflow automation
      </p>
      <h2 className="mt-2 text-[38px] leading-[0.95] text-text-primary sm:text-[48px]">
        <span className="block">Build</span>
        <span className="block">reliable</span>
        <span className="mt-1 inline-block bg-surface-cta-primary px-1.5 whitespace-nowrap">
          automation agents
        </span>
      </h2>
      <p className="mt-4 max-w-[56ch] text-[14px] leading-[1.45] text-text-secondary">
        Workflow automation includes dozens of steps across systems and
        frameworks. Trace the full execution, track cost at every step, and
        continuously include human expertise to improve your system.
      </p>
      
        <a
          href="/cloud"
          className="border border-line-cta bg-text-primary px-3 py-1.5 font-mono text-[11px] uppercase tracking-[0.06em] text-surface-bg transition-opacity hover:opacity-90"
        >
          Start free
        </a>
        <a
          href="/talk-to-us"
          className="border border-line-structure bg-surface-bg px-3 py-1.5 font-mono text-[11px] uppercase tracking-[0.06em] text-text-secondary transition-colors hover:border-line-cta hover:text-text-primary"
        >
          Talk to sales
        </a>
        <a href="/docs/observability/get-started" data-use-case-action="quickstart" className="inline-flex items-center py-1.5 text-[12px] text-text-tertiary underline decoration-line-structure underline-offset-4 transition-colors hover:text-text-primary hover:decoration-text-tertiary focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring sm:ml-2">Tracing quickstart →</a>
      
    

    
      
- ingest_pdf
- ocr_pages ×14
- extract_fields
- reconcile_retry ×3 ($0.061)
- post_to_erp

one retry step · 66% of run cost

    

  </section>

  <section data-use-case-section="benefits" className="border-t border-line-structure px-5 py-14 sm:px-10 sm:py-20">
    
      <h2 className="text-[34px] leading-[1] text-text-primary sm:text-[46px]">
        Trace the full execution across systems
      </h2>
    

    
### Easily debug your agent executions

Trace every step, tool call, and input and output your agent produces. Add deep links to and from your product to easily debug traces. Find failure modes and bottlenecks across executions, then prioritize issues by how often they happen so you fix what actually matters.

![Agent execution trace showing model turns, tool calls, approval waits, and failed dashboard-widget operations](/images/workflow-automation/execution-trace.png)

[Tracing overview](/docs/observability/overview)

### Get a full view on cost at every step

See the cost of an execution as a whole and a granular view of each sub-step. Swap models and inspect quality so you can manage the cost of the system without guessing.

![Agent execution trace with the total run cost and individual model-step costs highlighted](/images/workflow-automation/step-costs.png)

[Cost tracking](/docs/observability/features/token-and-cost-tracking)

### Leverage human expertise to improve your system

Let engineers and domain experts annotate data in Langfuse and mark when the system went wrong. Log corrections from human-in-the-loop setups to build ground-truth data at scale.

![Human annotation of an incorrect tool query, with a written explanation and a failed assessment](/images/workflow-automation/human-review.png)

[Annotation queues](/docs/evaluation/evaluation-methods/annotation-queues)

  </section>

  <section data-use-case-section="customer_stories" className="border-t border-line-structure bg-surface-1 px-5 py-14 sm:px-10 sm:py-18">
    <h2 className="mt-2 text-center text-[44px] leading-[1] text-text-primary">
      Teams improving workflow automation agents on Langfuse
    </h2>
    
### [Hugging Face](/users/hugging-face)

> “Looking at the traces in Langfuse, I saw that running a single LLM call wasn't reliable enough.”
>
> — Niels Rogge, Machine Learning Engineer, Hugging Face

### [Ravenna](/users/ravenna)

> “It's probably multiple orders of magnitude faster than it would be without a tool like this.”
>
> — Kevin Coleman, Co-founder, Ravenna

### [Merck](/users/merckgroup)

> “Langfuse enables us to track every prompt, response, cost, and latency in real time, turning black-box models into auditable, optimizable assets.”
>
> — Walid Mehanna, Chief Data & AI Officer, Merck

  </section>

  <section data-use-case-section="integrations" className="border-t border-line-structure px-5 py-8 sm:px-10 sm:py-10">
    
## Any system, any workflow

Trace n8n, Temporal, LangGraph, and no-code builders through native integrations or OpenTelemetry — nothing else in your stack changes.

### Workflow platforms

- [n8n](/integrations/no-code/n8n)
- [Temporal](/integrations/frameworks/temporal)
- [Restate](/integrations/frameworks/restate)
- [Zapier](/integrations/other/zapier)

### Agent frameworks

- [LangGraph](/integrations/frameworks/langgraph)
- [LangChain](/integrations/frameworks/langchain)
- [CrewAI](/integrations/frameworks/crewai)
- [OpenAI Agents](/integrations/frameworks/openai-agents)

### No-code builders

- [Dify](/integrations/no-code/dify)
- [Flowise](/integrations/no-code/flowise)
- [Langflow](/integrations/no-code/langflow)

  </section>

  <section data-use-case-section="resources" className="border-t border-line-structure px-5 py-14 sm:px-10 sm:py-16">
    <h2 className="mt-2 text-center text-[40px] leading-[1] text-text-primary">
      Learn from real automation examples
    </h2>
    <p className="mx-auto mt-2 max-w-[60ch] text-center text-[13px] text-text-tertiary">
      Start tracing a full application, evaluate structured extraction, and
      study patterns for multi-step LLM systems.
    </p>
    
      
- [Get started with tracing](/docs/observability/get-started): Ingest your first trace and see a full application run in Langfuse.
- [Structured output extraction cookbook](/guides/cookbook/example_structured_output_extraction): Evaluate document and claim extraction per field, then improve the pipeline with experiments.
- [Observability in multi-step LLM systems](/blog/2024-10-observability-in-multi-step-llm-systems): How to debug, evaluate, and test applications that chain many LLM and tool steps.

    
  </section>

  <section data-use-case-section="closing_cta" className="border-t border-line-structure bg-surface-1 px-5 py-10 sm:px-10 sm:py-12">
    
      
## Ready to build reliable automation agents?

Use traces, cost analytics, and human review to observe and improve every workflow run with Langfuse.

- [Start free](/cloud)
- [Documentation](/docs)
- [Talk to sales](/talk-to-us)

    
  </section>

<!-- agent-instructions -->

---

## Agent Instructions

This page is part of the [Langfuse](https://langfuse.com) documentation, published as plain Markdown for AI agents. Every page is available as Markdown by appending `.md` to its URL, or by sending an `Accept: text/markdown` header. This page: `https://langfuse.com/workflow-automation.md`.

### Querying these docs

If the answer is not on this page, query the documentation instead of guessing:

- **Semantic search** across all Langfuse docs, returning an answer with the relevant pages and excerpts. Ask a specific, self-contained question:

  ```bash
  curl -sG "https://langfuse.com/api/search-docs" --data-urlencode "query=How do I trace a LangGraph agent?"
  ```

- **Index of every page**: <https://langfuse.com/llms.txt>, with per-section indexes [llms-docs.txt](https://langfuse.com/llms-docs.txt), [llms-integrations.txt](https://langfuse.com/llms-integrations.txt), and [llms-self-hosting.txt](https://langfuse.com/llms-self-hosting.txt).

### Before writing Langfuse code

- **Install the [Langfuse Agent Skill](https://langfuse.com/docs/api-and-data-platform/features/agent-skill).** It encodes Langfuse's own best practices for instrumentation, prompt management, and evaluation, and materially improves results.
- **Read [What does a good trace look like?](https://langfuse.com/docs/observability/best-practices.md)** before instrumenting an application.
- **Verify endpoints, parameters, and response fields** against the [API reference](https://api.reference.langfuse.com) instead of inferring them from code examples.
- **Use the [Langfuse CLI](https://langfuse.com/docs/api-and-data-platform/features/cli)** (`npx @langfuse/cli api <resource> <action>`) to read or write traces, prompts, datasets, and scores from the terminal.

Found an error in these docs? Please open an issue at <https://github.com/langfuse/langfuse-docs/issues>.