Skill Detail

Run multi-surface coding-agent workflows with Kilo Code

Use Kilo Code as an open-source coding agent across VS Code, JetBrains, CLI, cloud runs, and trusted CI workflows with custom agents, MCP tooling, model switching, and review/debug modes.

Developer ToolsCustom Agents
Developer Tools Custom Agents Published
⭐ 26.3k GitHub stars ⬇ 248.3k/wk npm
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill run-multi-surface-coding-agent-workflows-with-kilo-code Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
At a glance
Tools required
Kilo Code VS Code extension, Kilo Code JetBrains plugin, or @kilocode/cli; local repository checkout; configured model access or Kilo account; optional MCP servers; optional trusted CI environment
Install & setup
Install the VS Code extension from the marketplace, install the JetBrains plugin, or install the CLI with npm install -g @kilocode/cli or brew install Kilo-Org/tap/kilo. Then run Kilo in a project directory and configure model and optional MCP access as needed.
Author
Kilo Code contributors
Publisher
Open Source
Last updated
Jul 17, 2026
Quick brief

Use Kilo Code when an operator needs the same repository-aware coding-agent workflow across an IDE, terminal, cloud run, or trusted CI job. The operator installs the VS Code extension, JetBrains plugin, or CLI, opens a project, chooses a task mode such as Plan, Code, Debug, Ask, or Review, connects model access and optional MCP servers, then runs implementation, debugging, review, or automation tasks with project context. Invoke this workflow instead of ordinary IDE use or one-off chat when work needs repeatable task modes, local file awareness, terminal execution, model switching, and auditable repository changes. Keep the scope bounded to operating Kilo Code as a coding-agent runtime for software tasks: planning, code edits, review passes, debug traces, terminal runs, and controlled automation. Inputs are a repository, model credentials or Kilo account access, and any optional MCP tools. Outputs should be reviewed diffs, plans, logs, or CI-ready automation steps, not an unbounded listing of every Kilo platform feature.