AI Engineering

Guides founders to compress project assumptions into a minimal market test with 10-minute start, 24-hour deliverable, 7-day metric, cost cap, and data collection to get real evidence.

Helps entrepreneurs reframe chaotic narratives and anxiety into a single, actionable business bottleneck by separating facts, interpretations, emotions, and assumptions, and identifying the earliest missing link in the revenue chain.

An AI coaching system for Chinese-speaking entrepreneurs using Claude Code. It turns vague business ideas into clear problems, pinpoints the first unvalidated step in the customer-to-revenue chain, and recommends a single low-risk experiment to test with real users. Suitable for side projects, content businesses, and early-stage ventures.

Entry point to the TC knowledge base, used to search TC methods, historical expressions, and external theories. It also manages knowledge atoms, connects user data sources, and validates knowledge integrity.

Monitors AI lab engineering blogs to filter out product announcements and surface only in-depth posts on autonomous agent architecture, memory, and orchestration. Use when you need a curated weekly digest of fundamental AI agent research.

A local-first knowledge distillation assistant that extracts methodologies, aesthetics, and technical insights from code repos, websites, videos, images, and documents. Use it for project collection, competitor research, style extraction, product planning, and implementation roadmaps.

Lore provides long-term project memory for AI coding agents, capturing architecture, decisions, and conventions in Markdown files. Use it when you need to record, recall, or synchronize project context across coding sessions and tools.

Simulates a film director’s creative team to transform a brief scene description into cinematic English prompts and negative prompts for AI image models like Midjourney, Flux, and SDXL. Use when you need film-quality text-to-image generation.

Transforms simple scene descriptions into professional cinematic image-generation prompts by simulating a film production team's workflow. Ideal for generating photorealistic, emotionally compelling images with a cinematic aesthetic.

Implements the shortest correct code that satisfies a locked PLAN.md, then verifies and updates the checklist. Used after planning and human approval to execute a coding task with minimal scope.

A global progress cockpit for AI coding projects using plans.md, progress.md, and findings.md to track goals, phases, and hardpoints. Use it to reprioritize tasks, resume after context clears, and keep the big picture in focus.

Routes image editing tasks like inpainting, background removal, upscaling, outpainting, and photo restoration to the most suitable engine. Use for precise surgical edits on existing images.