Build LLM Tool — Testing skill for Claude Code
Build the LLM chat-history tool app (Ktor + Koog + SQLite) per spec: invoke agents strictly in order, run acceptance checks, report.
How to install Build LLM Tool
Installs to ~/.claude/commands/p1nkyswear-chat-history-harness-build-llm-tool.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/P1nkySwear/chat-history-harness/HEAD/.claude/commands/build-llm-tool.md -o ~/.claude/commands/p1nkyswear-chat-history-harness-build-llm-tool.md Restart Claude Code, or start a new session, for it to be picked up.
What Build LLM Tool does
description: Build the LLM chat-history tool app (Ktor + Koog + SQLite) per spec: invoke agents strictly in order, run acceptance checks, report allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Agent, Skill, WebFetch, WebSearch argument-hint: "[optional: path to a spec file]"
/build-llm-tool — build the LLM chat-history tool app
You are the **orchestrator**. You do not write application code yourself: you invoke agents sequentially via the **Agent** tool and control their output.
Alternatives in Testing
- Debug — You are tasked with helping debug issues during manual testing or implementation 10k ★
- Agent Skills Spec — A skill is a folder of instructions, scripts, and resources that agents can discover and load dynamically to p 1.9k ★
- Mobile Scan — Quick QA scan of an existing mobile app 107 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Phased Engineering Pipeline
Claude Code skill for building software with AI agents: seven roles, vertical-slice phases, deterministic chec
Autoconst Spec Compliance Claude
Cross-check a Revit model against its written CSI construction spec. Auto-indexes the spec into SQLite, auto-e
Q Validate
Full validation of Spec-Kit-Plus workflow - checks order, artifacts, and detects violations
Conversational
Spec a conversational UI — voice assistant, chatbot, AI chat (LLM), or live-agent chat. Modality, persona, int
Ragagentsv1v2
RAG chat agent for your own documents, built in two stages. V1 answers with local models only (hybrid search,
Review Spec Implementation
Verify that the implemented code matches its feature spec - checks every acceptance criterion, data model, and
Related Agents
Feature Crew Reviewer
Code reviewer. Spawned with fresh context — no memory of build. Reads spec.md, design.md, ADRs, and the diff o
QA Documenter
Oracul QA documenter. Writes the release QA pack (test plan, acceptance report, how-to-run) strictly from real
Wd Reviewer
Use this agent when a work-package PR is ready for review, before merge. It reviews the diff strictly against