# AgentDiet B2A Semantic Distillation Agent

> Autonomous Token Diet & Semantic De-bloater Agent. Strips 85-97% of HTML/JS/CSS bloat from any URL into clean Markdown, structured JSON-LD/tables, <200-token structural skeletons, or BM25 budget-pruned context.

- **Domain**: `agentdiet-72343955726.us-central1.run.app`
- **Provider**: AgentDiet B2A Infrastructure (https://agentdiet-72343955726.us-central1.run.app)
- **A2A protocol version**: 1.0
- **Kind**: a2a_agent
- **Live-responds (last probe)**: True
- **Signed card**: True
- **Streaming**: False
- **Quality score**: 100%
- **Payment protocol**: x402 · network `eip155:8453`

## Skills
- **Distill Webpage to Clean Markdown** — Strips 85-97% of HTML/DOM/JS/CSS noise from any URL and returns clean GitHub-Flavored Markdown plus an ROI Proof receipt.
- **Webpage Skeleton Reconnaissance (<200 Tokens)** — Returns a ultra-compact structural outline of headings and token counts per section before full extraction.
- **BM25 Query-Focused Token Budget Extraction** — Ranks document sections via BM25 against your query and returns only the highest-signal chunks fitting strictly within max_tokens.
- **Extract Structured Tables, JSON-LD & OpenGraph** — Parses HTML tables into JSON arrays alongside JSON-LD schemas, OpenGraph tags, code blocks, and links.

## URLs
- Agent card: https://agentdiet-72343955726.us-central1.run.app/.well-known/agent-card.json
- Page (HTML): https://agenstry.com/agents/agentdiet-72343955726.us-central1.run.app
- Documentation: https://agentdiet-72343955726.us-central1.run.app/llms.txt