Previews
Deeper walkthroughs of the flagship apps, tools, and AI agents — what each one does, how it works, why I built it, and where it still falls short.
One terminal for all agents.
A build loop that checks what it shipped against what it promised.
Agent Rally Point Run Claude, Codex, and any agent with a shared coordination record.
Know your stack before you change it.
Secrets Vault Your agent can use the key without ever seeing the value.
One agent spec drives the canvas, the run, and the package.
Capture the running app on any platform, then ship the demo without opening an editor.
Design it, build it, and prove it — in one loop.
Agent Astronomer Search your Claude Code library before you build the same thing twice.
Turn scattered AI coverage into a connected, cited knowledge graph.
Deep research that plans, searches, and writes without a single byte leaving your machine.
Describe what you want to build get docs that actually agree with each other.
Direct the fleet instead of babysitting it.
One parser for every document that comes out ready for a model to read.
One question runs an entire research team, phase by phase.
See the design before it gets built.
Route the request, converge on a decision, and publish a spec a builder can execute.
One choice sends you down the path built for your actual job.
Nothing in that category yet.
RossLabs AI Plugins
18 open-source plugins are available for Claude Code; 17 are also available through the Codex marketplace mirror.
Install
# Add the marketplace — owner/repo, not the GitHub web URL
claude plugin marketplace add tyroneross/RossLabs-AI-Toolkit
# Install any plugin from it
claude plugin install build-loop@rosslabs-ai-toolkit# Add the Codex marketplace mirror
codex plugin marketplace add tyroneross/RossLabs-AI-Toolkit --sparse .agents/plugins
# Refresh installed and enabled plugins
codex plugin marketplace upgrade ross-labs-local