# AI use cases for FlowX (AIpRoBuilder/FlowX) Repo: [AIpRoBuilder/FlowX](https://github.com/AIpRoBuilder/FlowX) | Use cases: 3 | Page: https://dedupedai.com/use-cases/aiprobuilder-flowx/ FlowX is an open-source MCP server and Python SDK that turns plain-language process descriptions into stored, runnable AI agent workflows. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Build a support ticket workflow from a chat request: https://dedupedai.com/use-cases/aiprobuilder-flowx/build-support-ticket-workflow-from-chat/ - Refine an existing workflow with a follow-up change: https://dedupedai.com/use-cases/aiprobuilder-flowx/refine-workflow-with-python-sdk/ - Delegate workflow generation to another agent over A2A: https://dedupedai.com/use-cases/aiprobuilder-flowx/delegate-workflow-generation-over-a2a/ ## How to build a support ticket workflow from a chat request with FlowX Repo: [AIpRoBuilder/FlowX](https://github.com/AIpRoBuilder/FlowX) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Describe a process in plain language and get a stored, runnable FastAPI workflow. **Who it is for:** A support or operations lead who wants a repeatable ticket triage process, and an agent client already connected to a local MCP server. **The problem:** A one-off chat answer about ticket triage is lost after the conversation and cannot be rerun. **The outcome:** A reusable workflow on disk under workspace/workflows/ that classifies a ticket, assigns a priority, and drafts a reply. ### Steps 1. Run `python3.10 -m venv .venv && source .venv/bin/activate` in the FlowX folder. 2. Run `python -m pip install -e .` 3. Copy the example settings with `cp .env.example .env`, then set FLOWX_LLM_PROVIDER, FLOWX_LLM_MODEL, FLOWX_LLM_API_KEY, and FLOWX_DEFAULT_WORKSPACE. 4. Register flowx-mcp in your MCP client using the config example in the README, with the absolute path to the executable. 5. Ask the agent: Use flowx to build a workflow that accepts a support ticket, classifies its category, assigns a priority, and drafts a reply. ### Prompt to paste into your agent ```text Use the FlowX MCP server in the repo at to build a workflow that accepts a support ticket, classifies its category, assigns a priority, and drafts a reply. Report the workflow ID and the paths of the generated workflow.json and node files. Verify that the files exist under the FLOWX_DEFAULT_WORKSPACE folder. ``` ### Caveat Generated workflows depend on the LLM provider and model you configure, so review the generated node code before relying on its output. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/aiprobuilder-flowx/build-support-ticket-workflow-from-chat/ ## How to refine an existing workflow with a follow-up change with FlowX Repo: [AIpRoBuilder/FlowX](https://github.com/AIpRoBuilder/FlowX) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Apply a targeted change to one node or the whole workflow without rebuilding everything. **Who it is for:** A developer who has a generated workflow and needs to add a step, such as input validation or a summary, without starting over. **The problem:** Re-prompting from scratch regenerates work that was already correct and makes changes harder to control. **The outcome:** An updated workflow in which only the affected nodes change, rerun as a service through the SDK. ### Steps 1. Run `python -m pip install -e .` in the FlowX folder after completing the .env setup. 2. Create a Python script that opens `FlowXClient(workspace="./workflows")` and calls build_workflow_json with your process description. 3. Call list_nodes on the workflow ID to get each node's step_id and node_id. 4. Call evolve(workflow_id, "Validate empty input.", target=nodes[0]["step_id"]) to change one node, or evolve without a target to let the engine decide. 5. Call run_service(workflow_id) to start or restart the workflow as a service. ### Prompt to paste into your agent ```text Using the FlowX Python SDK in the repo at , open the workflow , list its nodes, and apply the change: Validate empty input. to the node . Report the list of nodes before and after, and confirm that run_service starts the workflow. ``` ### Caveat The SDK examples in the README are the reference for the calls shown here, and the walkthrough scripts are linked from the README rather than reproduced in full. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/aiprobuilder-flowx/refine-workflow-with-python-sdk/ ## How to delegate workflow generation to another agent over A2A with FlowX Repo: [AIpRoBuilder/FlowX](https://github.com/AIpRoBuilder/FlowX) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Let a separate agent send a text request and receive generated workflow artifact paths. **Who it is for:** A platform engineer whose orchestrating agent needs to hand off workflow creation to a separate service on the same machine. **The problem:** Each agent builds its own workflow logic, so generated processes are not stored or shared in one place. **The outcome:** An A2A task that returns a completed or failed result along with the generated artifact paths. ### Steps 1. Complete the venv and `python -m pip install -e .` setup from the Installation section. 2. Start the server with `flowx-a2a --host 127.0.0.1 --port 8001`. 3. Open http://127.0.0.1:8001/.well-known/agent-card.json to confirm the agent card is served. 4. Send an A2A text message to the /a2a endpoint with the process description using the message/send method. 5. Call tasks/get to read the task result and confirm the status is completed or failed. ### Prompt to paste into your agent ```text Use the FlowX A2A endpoint at http://127.0.0.1:8001/a2a to send a text message describing with message/send. Poll tasks/get until the task finishes. Report the task status and the generated artifact paths, and verify that the task is completed only if every stage succeeded. ``` ### Caveat The A2A adapter currently exposes non-streaming workflow building only, so a client has to wait for the task to finish rather than receive progress updates. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/aiprobuilder-flowx/delegate-workflow-generation-over-a2a/