A chat model answers questions. A coding agent works inside a software environment. That difference sounds small until the agent starts reading files, editing code, running tests, opening browsers, creating commits, and explaining what changed.

Claude Code, Codex, and Google Antigravity are examples of agentic coding tools. They do not replace local LLM runtimes, but they show what happens when a model is wrapped in tools, permissions, project context, and verification loops.

Mental model: model plus tools plus workspace

A coding agent is not just a model. It is a system:

PartRole
ModelReasons, plans, writes, reviews, explains
WorkspaceFiles, repository history, project conventions
ToolsSearch, read, edit, shell, browser, GitHub, MCP connectors
InstructionsUser prompt, project rules, skills, style guidance
PermissionsWhat the agent can read, write, run, or access
VerificationTests, builds, screenshots, logs, reviews

The Agentic Loop

Observe
Read context
Decide
Plan action
Act
Call tool
Verify
Check result
This is why an agent can be more useful than a raw local chat model for coding. It can act, observe, and revise.

Claude Code

Anthropic describes Claude Code as an agentic coding tool that lives in the terminal. Its official docs describe capabilities such as navigating a codebase, editing files, running commands, creating commits, and using MCP for external data or custom tooling.

Beginner framing:

Good forWatch out for
Terminal-first coding helpIt uses hosted Claude models by default
Explaining and modifying existing projectsCommand permissions matter
Scriptable workflowsRequires account/auth setup
MCP-connected toolsMCP servers expand access and risk

Official docs to verify:

Codex

OpenAI positions Codex as a coding agent for building, reviewing, debugging, and automating software work. The OpenAI developer site includes Codex use cases such as code review, frontend work, app deployment, documentation updates, creating CLIs for Codex to use, and saving repeatable workflows as skills.

Beginner framing:

Good forWatch out for
Repository-aware coding and reviewProduct surfaces and available features change
Long-running or structured coding tasksPermissions and sandbox settings shape behavior
Repeatable workflows through skillsSkills need testing, not just nice wording
Tool and connector useConnectors can expose sensitive data

Official docs to verify:

Google Antigravity

Google Antigravity is described as an agentic development platform with an AI-powered IDE, Agent Manager, browser control, artifacts, and asynchronous agents. Its docs emphasize workspaces or projects, task-level artifacts, terminal/browser/editor surfaces, and configurable security settings.

Beginner framing:

Good forWatch out for
Agent-first software workIt is its own product environment
Multi-surface work across editor, terminal, and browserBrowser and terminal permissions matter
Artifacts and task-level reviewGenerated artifacts still need human review
Parallel or asynchronous agentsConcurrent agents can conflict without isolation

Official docs to verify:

Where local LLMs fit

Most polished coding agents currently rely on hosted frontier or coding-specialized models. Local LLMs can still fit into the workflow:

Local useWhy it helps
Private first passSummarize or draft before sending selected context to a hosted agent
Offline code explanationAsk questions about snippets locally
Local evaluationTest prompts, tools, and structured outputs cheaply
OpenAI-compatible endpointSome tools can point to local runtimes if they support custom base URLs
Hybrid fallbackLocal for routine work, hosted agent for difficult edits

The limiting factor is not only model intelligence. Coding agents need reliable tool use, long context management, strong instruction following, and good recovery from errors.

Key papers on tool use, agentic systems, and code generation

Practical workflow

  1. Use a local model to explore ideas, summarize files, or draft tests.
  2. Use a coding agent when you need repository edits, terminal runs, or browser verification.
  3. Keep permissions narrow at first.
  4. Read the diff before trusting the result.
  5. Ask the agent to explain tests and evidence, not just claim success.
  6. Turn repeated workflows into skills, rules, or project instructions.

What to verify because tools change

Verify current installation methods, supported operating systems, model defaults, MCP support, permission modes, and data-use policies from official docs before publishing exact setup instructions. Coding-agent products evolve quickly.

The practical lesson: local LLMs are engines. Coding agents are work environments. The best workflow often combines local experimentation with agentic tools that can actually inspect, edit, and verify a project.