agent-platform-railway

Add a new agent to this AgentOS. Runs guided discovery or takes a concrete idea, then generates agents/slug.py, registers it in app/main.py, adds quick prompts, restarts the container, and smoke-tests it live. Use whenever the user wants to add or create a new agent.

Run the eval suite (python -m evals), diagnose every failure, fix what's in scope, and loop until all cases pass. Use when evals are failing, or when the user wants to run, diagnose, or repair the eval suite.

User-driven loop to change an existing agent in this AgentOS — add a tool/MCP server/toolkit, add a capability (knowledge base, memory, sub-agent, scheduled task), refine its instructions, or fix a specific known bug, verifying each change against the live container. Use whenever the user names a concrete change to an agent. For autonomous hardening with no specific change in mind, use improve-agent.

Autonomous hardening loop for an existing agent — derive probes from the agent's INSTRUCTIONS, run them against the live container, judge responses, edit the agent file, and re-probe until it reliably does what its instructions say. No user input needed. Use to harden an agent against its stated intent; to make a concrete change instead, use extend-agent.

Repo-wide drift sweep for public-readiness — diff docs against code, confirm every agent is registered and reachable, every env var documented, every doc path exists, and scripts behave as advertised; auto-fix mechanical drift and flag the rest. Use before a public release or after a refactor.