What problem does it solve? Adding a new agent to an AgentOS platform normally requires writing the agent file, registering it in the app entrypoint, updating the manifest, restarting containers, and verifying it works — a multi-step process prone to import errors and misconfiguration. This Skill automates that entire workflow, including building knowledge-only product agents grounded in a product's documentation. ## Core Features & Use Cases - Guided agent creation: Runs discovery or takes a concrete idea, then generates agents/slug.py, registers it in app/main.py, adds a manifest entry with quick prompts, restarts the container, and smoke-tests the result. - Product agent pattern: Ingests a product's website or docs into a product knowledge base and builds a knowledge-only agent with strict anti-hallucination guarantees (citation-only answers, grounded refusals). - Verified toolkit grounding: Searches Agno docs for toolkit imports and verifies every import inside the Docker container before writing code, preventing dead platforms from missing packages. - Use Case: A user says "build me an agent for Acme's docs" — the Skill ingests up to 50 documentation pages, generates the agent file, registers it, and validates it with covered, out-of-scope, and off-topic probes. ## Quick Start Ask the assistant to create a new agent for a named task or product, for example by saying "create an agent that reviews GitHub pull requests" or "build an agent for https://docs.example.com".
