| name | web-agent-declarative-framework |
| title | Building the Web for Agents: A Declarative Framework for Agentic Web Interactions |
| version | 0.0.2 |
| engine | skillxiv-v0.0.2-claude-opus-4.6 |
| license | MIT |
| url | https://arxiv.org/abs/2511.11287 |
| keywords | ["Web Agents","Declarative APIs","Tool Specification","Machine-Readable Contracts","Agent Integration"] |
| description | Enable safe and efficient AI agent interactions with websites via declarative HTML extensions—define agent-executable tools and context using <tool> and <context> tags instead of relying on brittle UI inference. |
Build Agent-Ready Websites with Declarative Tool Specifications
Web agents typically infer capabilities by analyzing human UI—button labels, form structures, navigation patterns. This approach is fragile: agents misinterpret intent, trigger unintended actions, and fail gracefully on novel UI layouts. The Agentic Web framework (VOIX) shifts responsibility to developers: specify what agents can do explicitly via declarative HTML extensions.
Rather than agents guessing "what this button does," developers declare tools through <tool> and <context> tags. This creates a machine-readable contract for agent behavior, enabling safe, predictable, and auditable agent interactions while preserving user privacy and control.
Core Concept
Current agent-web integration suffers from two problems:
- UI Brittleness: Agents infer capabilities from human-centric UI elements (button text, form labels); changes to UI break agents
- Intent Ambiguity: UI elements designed for humans (e.g., "Submit") don't clearly specify agent-executable intent
VOIX solves this with two declarative mechanisms:
<tool> tags: Developers explicitly declare actions agents can invoke, specifying inputs, outputs, and side effects<context> tags: Specify relevant state information agents need to make decisions (user preferences, session state, current form data)
Developers build the website normally; agents see enhanced, machine-readable augmentations. A hackathon study showed participants regardless of experience could rapidly build functional agent applications.
Architecture Overview
<tool> Element: Declarative action specification with name, description, parameters (type, required, validation), output format<context> Element: State declaration with visibility scope (user-visible, agent-only, public); prevents oversharing sensitive data- Tool Registry: Browser-side index of available tools per page; agents query registry rather than inferring from DOM
- Privacy Layer: Developers control which context is exposed to agents vs. humans; decouples conversational state from website state
- Validation & Execution: Agents receive declaratively-specified parameters; framework validates before execution
Implementation Steps
Step 1: Define Tools. Markup website actions with declarative tool specifications.
