Agent overview
Kodik includes an autonomous coding agent that reads and edits files, runs terminal commands, searches the codebase, fetches URLs, drives the Integrated Browser for visual checks, and calls MCP tools — with every action visible in the chat and requiring your approval (or running automatically if you have configured auto-approve).
The agent operates in one of five modes that determine its behavior and the tools it can use:
| Mode | Purpose | Edits files? |
|---|---|---|
| Code (default) | Executes tasks: reads, edits, runs commands | Yes |
| Plan | Creates an implementation plan without changes | No |
| Ask | Answers questions about the codebase | No |
| Debug | Hypotheses → instrumentation → minimal fix | Yes |
| Educator | Teaches with explanations and comprehension checks | No |
See Modes for full details.
The agent has a set of built-in tools:
- Reading and navigation —
read_file,glob,rg(ripgrep),codebase_search - Editing —
edit_file,edit_files,edit_notebook - Terminal —
shell(run commands),command_status(poll background processes) - Network —
web_fetch(HTTPS only) - Browser preview —
previewto open local pages, inspect rendered UI, take screenshots, read console output, and interact with the page - Utilities —
todo_write,memory,ask_questions,read_lints - Subagents —
sub_agentto delegate subtasks - MCP — any tools from connected MCP servers
The full tool reference is in Tools.
Key workflow capabilities
Section titled “Key workflow capabilities”Planning
Section titled “Planning”In Plan mode the agent explores the codebase and saves a structured plan via the generate_plan tool without making any changes. Switch to Code to start implementing. See Planning.
Checkpoints
Section titled “Checkpoints”Kodik automatically creates a checkpoint after each tool use, saving a snapshot of the workspace in a shadow Git repository. You can roll back to any checkpoint at any time. See Checkpoints.
Auto-approve
Section titled “Auto-approve”By default the agent asks for confirmation before every action. You can configure fine-grained permissions — allowing file reads, searches, specific command patterns, and more — while keeping control over edits. In Autopilot mode all actions are approved except commands on the blocked list. See Auto-approve.
Subagents
Section titled “Subagents”The agent can delegate isolated subtasks to child agents via the sub_agent tool. The child agent runs in its own context and returns results to the parent. See Subagents.
Browser preview
Section titled “Browser preview”In Code and Debug mode, the agent can use the Integrated Browser to verify browser-rendered work. It can start a local preview, take screenshots, read the console, resize the viewport, inspect elements, and interact with the page while you watch or take over. See Dev-server detection and Tools.
Reviewing changes
Section titled “Reviewing changes”All file edits appear in the changes panel in the chat. You can accept or reject each file individually and open a multi-file diff viewer. See Reviewing changes.