AI Integration

Enable LLMs and AI coding assistants to build high-performance reactive systems using Signals.dart.

The Signals ecosystem is designed to be highly accessible for Large Language Models (LLMs) and AI coding assistants. We provide unified context streams and modular developer Skills to give AI assistants full mastery over the framework's reactive paradigms.

Unified Context Endpoint#

Instead of manually feeding individual files into your AI chat or workspace, you can reference our single-source-of-truth endpoints. They contain the entire documentation set concatenated into a structured Markdown stream:

Endpoint	Contents	Best Use Case
https://dartsignals.dev/llms.txt	Full user guides and core reference documentation.	Adding project context to lightweight assistants.
https://dartsignals.dev/llms-full.txt	Complete, comprehensive reference manual (expanded details).	In-depth code generation and technical codebases.

Adding to Local Workspace Context#

If your AI coding assistant is scoped to your local project directory, you can download the endpoint directly to your workspace:

curl https://dartsignals.dev/llms.txt > signals.md

Once downloaded to your project root, you can reference the file (e.g. using @signals.md in cursor/copilot) to supply instant, highly relevant context.

Prepackaged AI Developer Skills#

To enable AI agents and coding assistants to write clean, idiomatic code for version 7 of the reactive framework, we package modular, high-fidelity developer Skills in the repository under the /skills folder.

These skills act as a system prompt extension that teaches the AI best practices, common gotchas (like the async tracking gap), and lifecycle optimization tracks:

signals-preact-dart: Deep knowledge of the core reactive primitives (signal, computed, effect, batch, untracked).
signals-dart: Best practices for async signal operations (FutureSignal, StreamSignal), reactive collections, writable computed signals (linkedSignal), and value semantics.
signals-flutter: Comprehensive guides on element-level reactive UI integration, high-frequency GPU rendering bypasses, lifecycle side-effects, and .watch(context) extensions.
signals-migration-6-to-7: Detailed instructions, patterns, and refactoring techniques to safely upgrade applications from signals v6.x to v7.0.0.

One-Command Skill Installation#

You can instantly install all available signals skills into your local AI workspace using the unified skills.sh registry CLI:

npx skills add rodydavis/signals.dart

This makes the skills immediately discoverable and usable by any editor-based or terminal-based AI assistant.