new-os-model-library

Orchestrates creation of a Griptape Nodes library wrapping an open-source model repository.

2|2|Updated Mar 26, 2025
One-click install
npx skills add https://github.com/griptape-ai/griptape-nodes-library-template --skill new-os-model-library-griptape-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: new-os-model-library
Source: https://github.com/griptape-ai/griptape-nodes-library-template/tree/main/.claude/skills/new-os-model-library
Command: npx skills add https://github.com/griptape-ai/griptape-nodes-library-template --skill new-os-model-library-griptape-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Building a Griptape Nodes library around an open-source model involves many manual steps: researching the model repo, scaffolding the library from a template, adding git submodules, configuring manifests, and implementing node classes. This Skill automates that entire pipeline end-to-end. ## Core Features & Use Cases - Phased Orchestration: Runs research, library setup, and node implementation as separate subagent phases with gate checks between each stage. - Spec-Driven Implementation: Generates a persistent spec file in .scratch/ that drives node definitions, dependencies, and library configuration. - User Approval Gate: Presents the proposed node list for explicit approval before writing any files. - Use Case: Given the SAM3 repository URL and a freshly templated library repo path, produce a complete Griptape Nodes library with registered nodes, manifest JSON, advanced library file, and updated pyproject.toml. ## Quick Start Run the new-os-model-library skill with the open-source model repo URL and the path to your library repo created from the Griptape Nodes template.

Frequently Asked Questions about new-os-model-library

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a Griptape Nodes library for an open-source model?

Provide the OS model repo URL and the path to a library repo already created from the griptape-nodes-library-template. The skill runs research, setup, and node implementation phases, producing registered nodes, a manifest JSON, and an advanced library file.

What naming convention does a Griptape Nodes library repo follow?

The repo directory must be named griptape-nodes-library-<dependency_name>, and the inner package directory griptape_nodes_library_<dependency_name>. The older griptape-nodes-<name>-library form is not supported.

Can I review the nodes before any files are written?

Yes. After the research phase, the skill presents the proposed node list with class names and descriptions, then waits for explicit approval. You can skip, rename, or rescope nodes before implementation begins.

Should griptape-nodes-engine be listed as a project dependency?

No. The engine is the host that loads your library, so listing it in [project] dependencies would install a second engine. It belongs in [dependency-groups] dev, with the required version recorded in the manifest's engine_version field.

What happens if a phase fails during library creation?

Fix the issue in the library repo and re-run the failed phase's subagent. The spec file persists in .scratch/, so the research phase does not need to be repeated when fixing setup or implementation phases.