skills.rest
keqingli1129
keqingli1129COMMUNITY
Explore all skills by this author →
GitHub

run-locally

Start the local agent server and test its HTTP API endpoints.

Updated May 10, 2026
Search Tags:#troubleshooting#api testing#local development#curl#agent server#mlflow evaluation#streaming requests
One-click install
npx skills add https://github.com/keqingli1129/agent-langgraph-one --skill run-locally-keqingli1129
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: run-locally
Source: https://github.com/keqingli1129/agent-langgraph-one/tree/main/.claude/skills/run-locally
Command: npx skills add https://github.com/keqingli1129/agent-langgraph-one --skill run-locally-keqingli1129

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you start the agent locally and quickly verify the API/UI behavior when you need to test changes or troubleshoot development issues.

Core Features & Use Cases

  • Start the agent server: Launch the local server and ensure it’s reachable at http://localhost:8000 for immediate manual testing.
  • Run with common server options: Use hot-reload, custom ports, or multi-worker execution to match your development or production-like workflow.
  • Test the API: Send streaming and non-streaming requests via curl to validate request/response formats and behavior.
  • Run evaluation and unit tests: Execute MLflow-based evaluation and pytest to confirm quality and prevent regressions.

Quick Start

Run the agent server on your machine by executing: uv run start-app.

Frequently Asked Questions about run-locally

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

FAQPage Schema
How do I start my agent server on localhost for local development?▼

To start local development of your agent server, execute the `uv run start-app` command to launch the application locally at http://localhost:8000 for manual testing and troubleshooting.

Can I test streaming API requests locally using curl?▼

Yes, you can test streaming and non-streaming API requests locally using curl to validate request and response formats against your agent's HTTP endpoints running on the localhost server.

What options are available when running a local agent server?▼

When running a local agent server, you can select development options such as hot-reload for immediate feedback, custom ports for routing, and multi-worker execution to match production-like workflows.

How do I run MLflow evaluation for my agent during local testing?▼

To run MLflow evaluation during local testing, execute the `uv run agent-evaluate` command to assess your agent's quality and prevent regressions before deploying changes.

Why is my local agent server not working or reachable on the port?▼

Local agent server issues often stem from port conflicts, missing experiments, or incorrect auth configuration; use the provided troubleshooting guidance to resolve these common local development blockers.

Related Skills

Explore More Software Engineering Skills→
theslashdojotheslashdojo

server

Start and configure the Dojo registry server for API, search, and graph endpoints.

Official
Intermediate
leary-pokenleary-poken
1

databricks-mlflow-evaluation

Run MLflow GenAI evaluate() to score agent outputs with configurable scorers.

Community
Advanced
timequitytimequity

openai-api

Integrate with the OpenAI REST API using direct HTTP calls.

Official
Intermediate
AnanyaDBJAnanyaDBJ
6

run-locally

Run an AI agent server and chat UI locally for development and testing.

Community
Intermediate