Search iThub

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

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

AI Coding Assistants & Agent Tooling

AI-powered tools that integrate directly into development environments to assist with code completion, generation, review and refactoring.

Choosing guide

62 results in AI Coding Assistants & Agent Tooling

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

Advertisements
Advertisements
Focused collections

Shortlists for AI Coding Assistants & Agent Tooling

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

Open-source CI/CD tools for engineering teams →

2 matching tools from this category

Open-source IT tools →

56 matching tools from this category

Self-hosted IT tools →

37 matching tools from this category

How to choose ai coding assistants & agent tooling tools

AI coding assistants and agent tooling help developers write, change and review code with language models, either inside an editor or as agents that work through a terminal. The category also includes the tooling around those agents: orchestrators that run several agents in parallel on isolated branches, usage and cost trackers, policy guards for shell commands, and session viewers. Reusable instruction packs for agents belong with skills and plugins, not here.

Read the guide

Who uses these tools

  • Software developers delegating routine changes to agents
  • Engineering leads reviewing agent output and spend
  • Security teams governing what agents may execute

Capabilities to compare

  • Parallel agent runs in isolated working copies
  • Diff review and pull request creation
  • Token usage and cost reporting per project
  • Pre-execution policy checks on commands and installs
  • Support for several agent and model providers

Operating considerations

  • Source code and secrets sent to model providers
  • Agents running shell commands with developer permissions
  • Usage-based billing that grows with parallel runs
  • Dependence on one agent vendor's session format

Selection checklist

  1. Try the tool on a real repository task and measure how much of the output you keep.
  2. Check which agents and model providers it supports, and whether you can bring existing subscriptions or keys.
  3. Review the permission model for file writes, shell commands and network access before enabling unattended runs.
  4. Confirm whether code, prompts and logs stay local or pass through a third-party service.

What belongs in this category: A product belongs here when it either performs coding work through a model or manages, observes or constrains the coding agents that do, rather than only supplying prompts or instructions for them.

AI Coding Assistants & Agent Tooling on iThub in figures

iThub lists 62 tools in AI Coding Assistants & Agent Tooling. 56 of 62 are open source, 37 of 62 can be self-hosted, 60 of 62 have a free, freemium or open-source pricing model and 27 of 62 offer an API.

Tools listed
62
Open source
56
Self-hostable
37
Free to start
60
With an API
27
Web app
28
Desktop app
52

Recently added: OpenComputer, Paseo, Free Claude Code, oh-my-hermes

Typical use cases

  • Running several agents on separate tasks in parallel branches
  • Tracking token spend across tools, models and projects
  • Blocking risky shell commands and secret access by agents
  • Reviewing agent changes as diffs before opening a pull request

Trade-offs to model

  • Editor-integrated assistants keep developers in control line by line, while autonomous agents cover larger tasks but need stricter review.
  • Local orchestrators keep code on your machines, whereas cloud workspaces simplify collaboration at the cost of sending repositories elsewhere.

Common mistakes

  • Letting agents merge their own changes without human review or a passing test suite.
  • Ignoring cost visibility until a month of parallel agent runs produces an unexpected invoice.
  • Granting agents the same credentials developers use, including production secrets stored in local configuration files.
Decision resources

Continue the technical evaluation

Use the resources that match your decision stage. Profile data and linked sources were refreshed through November 11, 2026.

Evaluation methodology · Corrections process