frontend-runtime-skills-contract

Community

Securely execute frontend runtime capabilities from tool calls.

Authorzxc1a1a1
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It prevents the frontend from executing or rendering unsafe, mismatched, or unregistered capabilities when it receives streamed tool-call data from agents.

Core Features & Use Cases

  • Runtime Capability Registry Contract: centralizes the mapping from toolName to component, behavior, risk level, and failure mode without binding to any specific agent.
  • Tool Call Arg Aggregation + Validation: aggregates fragmented args by toolCallId, parses JSON at TOOL_CALL_END, and validates against parametersSchema before executing.
  • Security-Aware Rendering and Degradation: enforces sandboxing rules (e.g., iframe for web_preview), disables reserved/disabled capabilities, and applies consistent fallback behavior (error cards, placeholders, or text fallback) to avoid white screens.
  • ToolResult Boundary Guidance: defines when ToolResult should or should not be returned, including safe error/error-card practices.

Quick Start

Register your frontend runtime capabilities in the Runtime Capability Registry keyed by toolName and then render only after the tool call ends, JSON parses, schema validates, and the registry status/risk policy allows execution.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: frontend-runtime-skills-contract
Download link: https://github.com/zxc1a1a1/Multi_Agent-AgentHub/archive/main.zip#frontend-runtime-skills-contract

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.