TandemKit
Dual-model planning, coding, and independent verification loop
All Skills in This Repository (3)
Pure Emerald Level Indicators
FlineDevevaluator
Verify Generator outputs against the Spec with Codex cross-checks.
FlineDevplanner
Investigates repositories and generates Spec.md to guide Generator and Evaluator workflows.
FlineDevgenerator
Implement mission specs, commit milestones, and signal the Evaluator.
Frequently Asked Questions
FAQPage SchemaHow to install TandemKit?▼
Run `npx skills add FlineDev/TandemKit --all -g -y` in your terminal to install all skills in this suite globally.
What does TandemKit do?▼
It runs a Planner, Generator, and Evaluator session that plan your feature, implement it, and independently verify the result in a loop until it passes. You only step in to approve the spec and review the final result.
Why use two models for code review?▼
A single session tends to approve its own work too early and miss real bugs. Pairing two independent models for planning and evaluation catches issues that one model alone overlooks.
Which project types does TandemKit support?▼
It includes evaluation strategies for iOS and macOS apps, Android and Flutter, web apps, CLI tools and libraries, and domain systems like tax or legal reasoning.
Do I need to supervise the coding loop?▼
No. After you approve the spec, the Generator and Evaluator coordinate autonomously through file-based signals, and you return only when the work passes verification.
Related Repositories in Software Engineering
View All in Software Engineering→
openclawopenclaw
Run a personal AI assistant across your devices and chat apps
obrasuperpowers
Gives coding agents a disciplined workflow from idea to merged code
facebookreact
AI agent skills for building, testing, and porting React core