Agent skill · software engineering · aradotso
kwcode-local-coding-agent
KWCode (天工开物) — a CLI coding agent optimized for local open-source models (8B-30B), featuring deterministic expert pipelines, BM25+AST code location, runtime debugging, and a self-improving flywheel — all running fully offline.
Why this skill is useful
Adds a comprehensive CLI tool for local coding with unique multi-task orchestration and expert pipelines that enhance coding capabilities.
What it needs
Requires docker, kwcode installed locally. About 6k tokens when loaded. Last updated 2026-07-09. 70 stars on the source repository.
What this skill does
KWCode Local Coding Agent Skill by ara.so — Daily 2026 Skills collection. KWCode is a CLI coding agent designed specifically for local open-source models (8B–30B parameters). Unlike cloud-first agents (Claude Code, Cursor), KWCode uses a deterministic expert pipeline where the LLM only classifies and generates — all routing, validation, and decision-making is handled by deterministic code. This lets small models succeed where they would otherwise hallucinate or loop. --- Installation Requirements: Python 3.10+, any OpenAI-compatible API (local or cloud). --- Quick Start On first launch, KWCode runs a setup wizard to configure your model connection. --- Configuration Connect to a local inference engine (Ollama example) Connect to a cloud API (no local GPU needed) Environment variable approach Recommended models by VRAM VRAM Model ------ ------- 4 GB gemma3:4b 8 GB qwen3:8b 16 GB qwen3:14b 24 GB+ qwen3:30b-a3b --- Key Commands All commands are entered inside the kwcode REPL (> prompt). Core task commands Planning & safety Shows execution steps and risk level (High / Medium / Low) before touching any file. Requires confirmation. Manually snapshot current project state. KWCode also auto-snapshots before each task. Restore to the last checkpoint if a task goes wrong. Multi-task orchestration (DAG) Interactive multi-task example: Search enhancement Without Docker, KWCode falls back to DuckDuckGo automatically. Statistics & flywheel Shows tasks completed, estimated time saved, and flywheel expert promotions. API management --- Project Configuration Files KWCode looks for these files in your project root and injects them as context: File Purpose ------ --------- KWCODE.md Project-level rules, conventions, coding standards. Injected per task type. PROJECT.md Auto-maintained project summary (Layer 1 memory). EXPERT.md Domain expert knowledge accumulated by the flywheel (Layer 2). PATTERN.md Recurring code patterns learned from your project (Layer 3). …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aradotso/kwcode-local-coding-agent