Menu
Convert the kimchi plugin into OpenCode and Codex compatible formats using the bundled converter CLI. Requires bun runtime.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | kimchi:convert |
| description | Convert the kimchi plugin into OpenCode and Codex compatible formats using the bundled converter CLI. Requires bun runtime. |
| argument-hint | [--to opencode|codex|both] [--output path] |
Convert the kimchi Claude Code plugin into OpenCode and Codex compatible formats.
Find these two absolute paths:
.claude-plugin/plugin.json. This is the kimchi plugin root.cli/ subdirectory. It contains src/index.ts and package.json.Both paths MUST be absolute. CWD varies across invocations.
which bun
If missing, tell the user to install bun: curl -fsSL https://bun.sh/install | bash
Check if node_modules/ exists in the CLI directory. If not:
cd <cli-dir> && bun install
Parse the user's arguments to determine the target format. Default is both.
OpenCode only:
bun run <cli-dir>/src/index.ts convert <plugin-root> --to opencode --output <output-path>
Codex only:
bun run <cli-dir>/src/index.ts convert <plugin-root> --to codex --output <output-path>
Both (default):
bun run <cli-dir>/src/index.ts convert <plugin-root> --to opencode --also codex --output <output-path>
Default output to .converted/ in the project root (the git repository root or the directory containing the kimchi plugin), not the current working directory.
If the user passes --output <path>, use that instead.
After conversion, report:
opencode.json, config.toml)commands/ directory in the converter sense — only skills/ and agents/. The commands array in output will be empty. This is expected; skills are kimchi's functional units.install command in the CLI is compound-engineering specific and not useful for kimchi.This command should be used to run the Kimchi planning pipeline through refinement, transforming a vague idea into a draft plan ready for cross-model analysis. Orchestrates 6 stages: clarify, requirements, research, generate, review, refine. Use --full-auto to also run beads + validate after manual revise/synthesize.
Use when executing a bead task. Detects orchestration mode (ACFS or GasTown) from .beads/manifest.yaml and enforces the appropriate execution discipline.
This command should be used to convert the final plan into standalone bead YAML task specifications for multi-agent execution. Ninth stage of the Kimchi planning pipeline. Produces .beads/ directory. Supports both ACFS and GasTown orchestration modes.
This command should be used to run the Kimchi self-improvement system. Observes execution outcomes, proposes incremental improvements to skills/validators/personas, validates against history, and applies with versioning.
This command should be used to clear the Kimchi working directory (.kimchi/) and start fresh. Preserves .beads/ directory. Use when starting a new planning session or recovering from a bad state.
This command should be used to validate bead YAML files for standalone executability. Runs 4 validators (context completeness, deliverables clarity, test specification, isolation) and enriches failing beads. Eighth stage of the Kimchi planning pipeline.