Kilo Code

Kilo Code is open-source VS Code agent in the Cline/Roo lineage, with an orchestrator mode and a CLI. We compare it in Coding agents. It is one of only 5 of 14 tools in its roster with Subagents. The trade-off is being marked No on Background agents, where Claude Code records Yes.

What is Kilo Code?

Kilo Code — Open-source VS Code agent in the Cline/Roo lineage, with an orchestrator mode and a CLI. Kilo Code is an open-source VS Code extension descended from the Cline and Roo Code lineage, merging features from both and adding an orchestrator mode that decomposes work across specialised sub-modes. It ships its own credit-based provider alongside standard BYOK, and has since added a CLI. It moves fast and inherits a mature extension architecture, but its commercial side is less documented than its rivals'. Open source under Apache-2.0.

We track it in 1 comparison — Coding agents — so every claim below is a cell in a table you can open and check rather than an impression. Across those rosters it sits against 16 other tools, and what follows is where it visibly separates from them.

Where it wins.

  • Subagents — Yes, which only 5 of 14 tools here manage.
  • Repo indexing — embeddings index, 3rd of 16.

Each of those is ranked against the whole roster on its category page, not against a hand-picked subset, so a first place here means first of everything we list.

Where it gives ground.

  • Background agents — No. Claude Code records Yes.
  • GitHub stars — 9,000, the weakest of the 9 here. Gemini CLI records 75,000.
  • Opens PRs — No. Claude Code records Yes.

None of these disqualify it on their own. They are the fields to check against your own requirements before you commit, because they are the ones where a competitor genuinely does better.

Provenance. 22 of 26 tracked fields carry a value for Kilo Code, and 2 of those cite a document you can open. Nothing here carries a verification date yet, which is itself a reason to check before committing. Every figure keeps its own provenance — measured by us, claimed by the vendor, inferred, or community-reported — and we would rather print a dash than a guess.

Its nearest neighbour in our data is Aider. Kilo Code is ahead on MCP (Yes against No) and Subagents (Yes against No). Aider takes Headless / CI (Yes against Partial) and GitHub stars (37,000 against 9,000). That pattern repeats across the rest of the roster — see Kilo Code alternatives for the other rivals, each compared the same way.

At a glance
Licence
Apache-2.0 (open source)
Fields we track
22 of 26

Kilo Code in Coding agents

Ranked against 17 tools across 26 sourced fields. Open the full Coding agents table.

A features-merged descendant of Cline and Roo Code that ships faster than either, with orchestrator mode as the genuine differentiator. The commercial side — credits, markup, roadmap — is the least documented of any open-source tool here, which is why this row carries nulls.

Where it lands in this roster
Subagents
Yes1st of 14
Community-reported
Background agents
No12th of 16
Inferred
GitHub stars
9,0009th of 9
Community-reported
Repo indexing
embeddings index3rd of 16
Community-reported
Opens PRs
No10th of 16
Inferred
BYOK
Yes1st of 16
Community-reported
MCP
Yes1st of 15
Community-reported
Entry price
$0 /seat/mo1st of 16
Inferred

When to use Kilo Code

Kilo Code is the right call in these situations, each one drawn from a field we actually record:

  • VS Code teams who want mode-based orchestration without leaving the IDE.
  • People who liked Roo Code and want a more actively developed descendant.
  • Subagents is your binding constraint. Kilo Code records Yes, which only 5 of the 14 tools in the Coding agents roster do. We define that field as can the agent spawn child agents with their own context window and toolset, and collect their results? The main lever for keeping long tasks inside a context budget.
  • Repo indexing is your binding constraint. Kilo Code records embeddings index, 3rd of 16 in the Coding agents roster. We define that field as how much structure the harness has about your code before the first token: nothing, live search, a parsed symbol map, a persistent vector index, or an index read alongside live search.

When not to use Kilo Code

Reach for something else when any of the following is a requirement rather than a nice-to-have:

  • Background agents. Kilo Code records No on the Coding agents table. Claude Code records Yes on the same field. If that is a hard requirement rather than a preference, start elsewhere.
  • GitHub stars. Kilo Code records 9,000, the weakest of the 9 tools in the Coding agents roster. Gemini CLI records 75,000 on the same field. If that is a hard requirement rather than a preference, start elsewhere.
  • Opens PRs. Kilo Code records No on the Coding agents table. Claude Code records Yes on the same field. If that is a hard requirement rather than a preference, start elsewhere.

We publish this block because a comparison that only lists what a tool is good at is marketing. Every figure above sits on the same page as its source, and the field definitions are on the category tables if you want to check how we measured them.

Tools compared alongside Kilo Code

Everything below shares at least one comparison category with Kilo Code, ordered by how much overlap there is. For the reasoning on each — which fields it wins, which it loses — see Kilo Code alternatives.
Minimal terminal pair-programmer with a tree-sitter repo map and automatic git commits.
Sourcegraph's opinionated agent with no model picker and an ad-supported free tier.
Open-source VS Code and JetBrains agent built around an explicit plan/act split.
Cursor
VS Code fork with a real codebase index, background cloud agents and a CLI.
Fully delegated cloud engineer: file a task in Slack or the web app, get a PR back.
DIY harnessbaseline
Your own loop over a model API with an agent SDK — the build-it-yourself reference point.

Recent Kilo Code changes

We have not logged a dated change for Kilo Code yet. The timeline fills in as pricing moves, features ship and things get deprecated.

Sources and gaps

What we don't know. 4 of the 26 fields we track for Kilo Code are still blank: Default model, Hooks, Terminal-Bench and Bench model. Those render as dashes rather than as zeroes or assumptions, because an empty cell and a bad cell are not the same thing and only one of them is honest. If you know any of these figures and can point at a document, tell us.

Every figure on this page traces back to a document you can open. Where a vendor claims a number we could not reproduce, the cell says so.