Configurable coding agent
How to use
Kilo Code.
Planning and completing repository tasks with specialized agents, configurable providers, project rules, granular tool permissions, and optional terminal automation.
Kilo Code is a open-source AI coding agent for VS Code and terminal workflows. Planning and completing repository tasks with specialized agents, configurable providers, project rules, granular tool permissions, and optional terminal automation. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.
AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.
Access & setup
Find, install, and sign in to Kilo Code
Get into the official Kilo Code experience with the right account and a setup you understand.
- 01
Start at https://kilo.ai/ and confirm the domain before entering account or payment information.
- 02
Availability: Kilo Code is available as a VS Code extension and CLI with coding, planning, debugging, custom-agent, autocomplete, MCP, and multi-provider workflows.
- 03
A supported editor or terminal, the official extension or CLI, and an approved built-in, cloud, gateway, or local model provider.
- 04
Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.
- 05
Run one low-risk test task before connecting sensitive files, repositories, or workspace data.
- Use official download pages.
- Review permissions during setup.
- Keep installers and applications updated.
First session
Your first useful Kilo Code session
Learn the interaction loop using a small task with a clear outcome.
- 01
Install the official extension or CLI and select an approved provider without storing credentials in the repository.
- 02
Open a clean branch and change consequential tools from allow to ask or deny before the first task.
- 03
Use a planning agent to inspect the request and agree on acceptance criteria before allowing edits.
- 04
Review commands, changed files, tests, costs, and the final diff before committing or shipping.
- State the outcome before the background.
- Provide the real source material.
- Review the result before expanding the task.
Quality control
Check the quality of Kilo Code output
Turn generated code into a change you would put your name on in review.
- 01
Restate the intended behaviour and the acceptance criteria before reading a single generated line.
- 02
Read the diff rather than the chat summary; the summary describes intent, the diff describes what actually changed.
- 03
Run the test suite, then write a test that would fail if the change were wrong, and confirm it passes for the right reason.
- 04
Check the edges the model tends to skip: error paths, null and empty cases, concurrency, and migration or rollback behaviour.
- 05
Trace every new dependency, network call, and file write, and confirm each one is genuinely required.
- Review the diff, never the description of the diff.
- A passing test suite proves the tests pass, not that the change is correct.
- Keep changes small enough that a human can actually review them.
Privacy & permissions
Use Kilo Code safely
Give the tool enough repository access to help, and no more than that.
- 01
Decide what the tool may read and what it may write before connecting a repository.
- 02
Scope tokens and integrations to the narrowest repository, branch, and permission set that still works.
- 03
Keep secrets out of the context window: use environment variables and secret stores, never inline credentials.
- 04
Require review before generated code touches authentication, payments, permissions, migrations, or deletion paths.
- 05
Audit what the integration retained after the session, and revoke access that is no longer needed.
- Kilo can read and edit files, execute shell commands, access external directories and the web, load skills and MCP tools, and launch other agents. Its smooth defaults allow many tools automatically, so review project and global rules before use, deny secrets and external paths, keep shell and side effects approval-gated, avoid autonomous mode on untrusted work, isolate parallel sessions, bound cost and retries, and independently verify every result.
- Follow your organisation's approved-use policy.
- Never treat fluent output as authorization to act.
Core workflows
Step-by-step ways to use Kilo Code for the work it does best.
Each workflow below has its own outcome, steps and review checkpoints.
Use planning and coding agents with explicit permission rules to move from issue to verified diff.
↓ 02 WorkflowCreate a least-privilege Kilo review agentBuild a reusable custom agent that can inspect code and run narrow diagnostics without editing or publishing changes.
↓Workflow 01
Complete a repository task safely with Kilo Code
Use planning and coding agents with explicit permission rules to move from issue to verified diff.
Outcome: Produce a scoped, tested change without letting permissive defaults or autonomous execution bypass review.
- 01
Create a clean branch, capture repository status and baseline checks, and inspect project guidance, generated files, and sensitive paths.
- 02
Review global and project permission rules, changing shell, edits, external directories, web access, MCP, skills, and subagents to ask or deny where appropriate.
- 03
Give the planning agent the outcome, acceptance criteria, relevant paths, non-goals, and validation commands, then approve a minimal implementation plan.
- 04
Use the coding agent interactively, reviewing every consequential action, changed file, dependency, command, test result, and attempt to broaden scope.
- 05
Run trusted checks independently, inspect the complete diff and usage cost, and merge or deploy only through the repository's normal human-approved workflow.
What good use looks like
- Audit defaults before the first prompt.
- Use interactive mode for consequential work.
- Keep release actions human-controlled.
Workflow 02
Create a least-privilege Kilo review agent
Build a reusable custom agent that can inspect code and run narrow diagnostics without editing or publishing changes.
Outcome: Standardize AI-assisted review while making its authority and evidence requirements explicit.
- 01
Define the review scope, severity rubric, safe evidence, exact diagnostic commands, maximum steps, provider, and output schema.
- 02
Create a project-scoped agent with read and search access while denying edits, broad shell, external directories, web tools, skills, MCP, and nested agents.
- 03
Allowlist only necessary diagnostic commands and paths, keeping environment files, credentials, installation, Git writes, pushes, and deployments blocked.
- 04
Test known defects, false positives, hostile comments, large diffs, generated files, missing dependencies, and requests to escape the permission boundary.
- 05
Version the agent configuration, require file-and-line evidence, manually verify serious findings, and review the rules whenever Kilo or the repository changes.
What good use looks like
- Deny editing and publishing.
- Allowlist exact diagnostics.
- Verify consequential findings manually.
Official references
Check the current product documentation.
Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.