Search iThub

Search tools, categories and collections. Type at least two characters.

Explore all tools
↑ ↓ to explore resultsEnter to open · Esc to close

Development & Code MCP Servers

Development & Code mcp servers for AI agents and assistants.

Choosing guide

13 results in Development & Code MCP Servers

Active Pro listings from this category appear first. Your sort applies within Pro and Free results.

Advertisements
Advertisements
Focused collections

Shortlists for Development & Code MCP Servers

Explore collections with tools relevant to this category, grouped by deployment, source model or workload.

Free IT tools →

8 matching tools from this category

Open-source IT tools →

12 matching tools from this category

Self-hosted IT tools →

5 matching tools from this category

How to choose development & code mcp servers tools

Development MCP servers give AI coding assistants structured access to repositories, local files, Git history, design files, browser automation and code analysis through the Model Context Protocol. Instead of pasting code into a chat, the client calls tools that list files, read diffs, open pull requests or drive a headless browser. This subcategory also covers context and memory helpers for coding agents, which sets it apart from infrastructure servers that manage clouds and databases.

Read the guide

Who uses these tools

  • Developers pairing with coding agents in an editor
  • QA engineers driving browser tests from an assistant
  • Maintainers triaging issues and pull requests

Capabilities to compare

  • Repository, issue and pull request operations
  • Filesystem access limited to allowed directories
  • Headless browser control and screenshots
  • Persistent memory or context compression for agents
  • Reference test servers for client developers

Operating considerations

  • Filesystem and shell tools can modify local code
  • Large tool output consumes the context window
  • Personal access tokens stored in client config
  • Local stdio process versus remote HTTP server
  • Isolating browser automation from logged-in sessions

Selection checklist

  1. Restrict filesystem servers to the project directories the agent needs and keep home folders and secrets out of scope.
  2. Use fine-grained, repository-scoped tokens for code host servers instead of a classic token with full account access.
  3. Measure how much context each tool response consumes and prefer servers that paginate or summarize large outputs.
  4. Try the server with the coding client you actually use, because transport and tool support vary between clients.

What belongs in this category: Servers listed here expose software development resources such as code, version control, design handoff, browser testing or agent memory to MCP clients, rather than cloud accounts, production databases or business applications.

Development & Code MCP Servers on iThub in figures

iThub lists 13 tools in Development & Code MCP Servers. 12 of 13 are open source, 5 of 13 can be self-hosted, 13 of 13 have a free, freemium or open-source pricing model and 4 of 13 offer an API.

Tools listed
13
Open source
12
Self-hostable
5
Free to start
13
With an API
4
Web app
2
Desktop app
5

Recently added: Kanzo, ARTEMIS, jscpd, Figma MCP

Typical use cases

  • Letting an agent open a branch and draft a pull request
  • Reproducing a front-end bug through scripted browser steps
  • Pulling design tokens and screenshots into a coding session
  • Keeping project decisions available across separate agent sessions

Trade-offs to model

  • Local stdio servers are simple and private but run with your user rights, whereas remote servers add OAuth and network latency.
  • Giving an agent many development tools increases what it can do, but large tool lists confuse tool selection and waste context.

Common mistakes

  • Pointing a filesystem server at the whole disk, which lets the agent read keys and dotfiles it never needed.
  • Installing many overlapping servers at once and then struggling to tell which tool the agent actually called.
Decision resources

Continue the technical evaluation

Use the resources that match your decision stage. Profile data and linked sources were refreshed through November 10, 2026.

Evaluation methodology · Corrections process