Kilo CodevsZoo Code
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Kilo Code
Kilo Code (Anaconda) · IDE extension
#26OSSMCP
Panel
6.55 spec wins
- Reliability
- 6.0
- Usefulness
- 7.2
- Cost
- 7.0
- Longevity
- 5.8
“A fork of Roo Code carrying a fork of OpenCode, acquired by Anaconda, so the dependency tree finally has a parent.”
Zoo Code
Zoo Code Org · IDE extension
#16OSSMCP
Panel
7.50 spec wins
- Reliability
- 7.0
- Usefulness
- 7.5
- Cost
- 8.5
- Longevity
- 7.2
“The original team wound down the extension to focus on Roomote, so the fork is now the product and the product is now a memory.”
Spec by spec
| Spec | Kilo Code | Zoo Code |
|---|---|---|
| Architecture | ||
| Category | IDE extension | IDE extension |
| Runs | local, cloud | local |
| Platforms | macos, linux, windows, web | macos, linux, windows |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | No |
| Browser control | YesA Playwright-backed browser tool that navigates, clicks, fills forms, drags and screenshots after each action; it needs a high-capability agentic model to work reliably. | NoThe documented tool set covers read, edit, execute, image and workflow tools; no browser tool is listed. |
| Sandboxed execution | YesCloud Agents run in isolated Linux containers, and the local CLI has an opt-in OS-level sandbox using macOS Seatbelt via sandbox-exec or Linux bubblewrap (not available on Windows). | No |
| Multi-agent orchestration | Yes | YesThe Orchestrator delegates subtasks with parallel coordination, per-subtask provider profiles and parent/child task recovery. |
| Headless / CI mode | Yes`kilo run --auto` auto-approves permissions for fully non-interactive CI/CD runs. | No |
| Models | ||
| Backbone | Claude, GPT, Gemini, Ollama, LM Studio | Claude, GPT, Gemini, Kimi, GLM, Grok, MiniMax, Zoo Gateway, Ollama, LM Studio |
| Bring your own model | Yes | Yes |
| Local models | Yes | Yes |
| Cost | ||
| Pricing model | mixed | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | Apache-2.0 |
| GitHub stars | 27,467 | 1,937 |
Which one would each critic pick
| Critic | Kilo Code | Zoo Code | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.3 | 7.8 | Zoo Code — Pick Zoo Code if you already know Roo Code's modes and want a version somebody is still maintaining; pick Cline if you would rather start from the tool with the largest crowd. |
| El Crítico | 6.0 | 7.0 | Zoo Code — Longer autonomous runs are permitted because a Destructive Command Guard blocks dangerous commands. That is a denylist, and a denylist is only as good as the last thing somebody thought of. |
| El Profesor | 5.5 | 7.0 | Zoo Code — Semble is described as codebase intelligence giving semantic search with no separate indexing step, which is a claim about when the work happens rather than about whether it happens. |
| La Inversora | 6.3 | 7.5 | Zoo Code — 1,793 stars in four months and a gateway selling pay-as-you-go credits with no subscription. That is a fork with a revenue line, which is rarer than the fork. |
| La Jefa | 6.8 | 7.5 | Zoo Code — Zero across sixty seats, and multi-root path controls plus per-mode server restrictions give me something to actually configure, which almost nothing free on this board does. |
| El Hacker | 7.3 | 8.5 | Zoo Code — Apache-2.0, my own keys at the provider's own rates, and Ollama and LM Studio are documented providers, so the whole loop runs on hardware I built if I want it to. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.