ClawHub

Materials Cli

v1.0.8

Render JSON schemas to images and generate schemas from prompts using declare-render and AI.

⭐ 0· 772·2 current·2 all-time
byWhite@cai-zhuo
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
ℹ
Purpose & Capability
Name/description (render schemas, generate via AI) align with the code: render, validate, and generate commands are implemented and use declare-render and OpenAI. One oddity: package manifests reference local file dependencies (e.g., "materials-agents": "file:../materials-agents" and declare-render via local paths) which is unusual for a published package and may indicate the repository was packaged from a monorepo rather than prepared for publishing. This is likely a packaging/installation issue, not a capability mismatch.
✓
Instruction Scope
SKILL.md and code only instruct the agent/CLI to load schema files, render them, validate them, or call an AI helper to produce schemas. The generate flow explicitly requires an OpenAI API key (or uses CLI flag), and no instructions request unrelated files, system credentials, or unknown external endpoints.
ℹ
Install Mechanism
Registry entry has no install spec (instruction-only), but the package includes source and lockfiles. Dependencies include native 'canvas' (node-canvas) which requires platform libraries and rebuild steps; the code surfaces helpful install guidance. The biggest oddity: package.json/lock reference file:../ local packages (declare-render, materials-agents). That makes typical npm/pnpm installation fragile or confusing when installing from a registry or an unknown source.
✓
Credentials
Declared required env is OPENAI_API_KEY (and the CLI also optionally reads OPENAI_MODEL and OPENAI_BASE_URL, which are reasonable). No other secrets or unrelated credentials are requested. The primary credential (OpenAI key) is appropriate for the generate command.
✓
Persistence & Privilege
Skill does not request always:true or any special permanent agent privileges. It does not modify other skills or system-wide configs. It runs as a normal CLI-style tool.
Assessment
This skill appears to do what it says — render/validate JSON schemas and optionally generate schemas using OpenAI. Before installing: (1) Be aware it depends on node and node-canvas (native system libs) — follow the provided install steps for your OS; (2) the package references local file dependencies (materials-agents, declare-render) which may not be available when installing from an external registry; the generate command will fail unless you install the materials-agents package or pass an API key and model via flags; (3) the generate flow sends your OPENAI_API_KEY to OpenAI (or to any base URL you set) — only provide a key you trust and consider using a key with limited scope or rate limits; (4) if installing code from an unknown source, review the materials-agents and declare-render packages (or run the CLI in an isolated environment) to ensure there are no unexpected network calls or behaviors. Overall: functionally coherent, but double-check packaging and the auxiliary packages before use.

Like a lobster shell, security has layers — review code before you run it.

latestvk97c4qrdrpghx9ja5mme1ymcqn82ey96

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

Binsnode
EnvOPENAI_API_KEY
Primary envOPENAI_API_KEY

Comments