Agent Margin Router

HTML to Markdown

html_to_markdown

Read-only

Convert any public web page into clean, LLM-ready Markdown: main content only (boilerplate, nav and ads removed), headings, lists, tables and optional links/images preserved. Trafilatura extraction with markdownify fallback, SSRF-guarded fetch, up to 3 MB HTML. Pay per call with USDC or USDT on Base via x402. 3 free calls per wallet.

Input Schema

TableJSON Schema

Name	Required	Description
`url`	Yes	Public page URL.
`wallet`	No	Your EVM wallet address (0x...). Unlocks the free tier (3 free calls per wallet).
`api_key`	No	Optional AMR enterprise API key.
`x_payment`	No	Optional signed x402 payment payload (base64 JSON) for USDC/USDT on Base.
`include_links`	No	Keep hyperlinks (default true).
`include_images`	No	Keep images (default false).

Output Schema

TableJSON Schema

Name	Required	Description	Default
`url`	Yes
`title`	No
`token`	No	Token actually used for payment (USDC, USDT or FREE).	USDC
`routing`	Yes	How the request was routed (provider, fallback, cache, latency).
`markdown`	Yes
`cost_usdc`	Yes	Price charged in stablecoin units (USDC/USDT 1:1); 0 on the free tier.
`final_url`	Yes
`truncated`	No
`char_count`	Yes
`fetched_at`	No	UTC timestamp when the data was fetched.
`word_count`	Yes
`include_links`	Yes
`include_images`	Yes
`extraction_method`	Yes
`freshness_seconds`	Yes	Age of the underlying data in seconds (0 = fetched live).
`x_payment_response`	No	Base64-encoded x402 settlement receipt (mirrors the X-PAYMENT-RESPONSE header). Only present when the call was paid with `x_payment`.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedSep 15, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "description": "Response for ``POST /html-to-markdown``.",
+  "properties": {
+    "char_count": {
+      "title": "Char Count",
+      "type": "integer"
+    },
+    "cost_usdc": {
+      "description": "Price charged in stablecoin units (USDC/USDT 1:1); 0 on the free tier.",
+      "title": "Cost Usdc",
+      "type": "number"
+    },
+    "extraction_method": {
+      "title": "Extraction Method",
+      "type": "string"
+    },
+    "fetched_at": {
+      "description": "UTC timestamp when the data was fetched.",
+      "format": "date-time",
+      "title": "Fetched At",
+      "type": "string"
+    },
+    "final_url": {
+      "title": "Final Url",
+      "type": "string"
+    },
+    "freshness_seconds": {
+      "description": "Age of the underlying data in seconds (0 = fetched live).",
+      "title": "Freshness Seconds",
+      "type": "integer"
+    },
+    "include_images": {
+      "title": "Include Images",
+      "type": "boolean"
+    },
+    "include_links": {
+      "title": "Include Links",
+      "type": "boolean"
+    },
+    "markdown": {
+      "title": "Markdown",
+      "type": "string"
+    },
+    "routing": {
+      "description": "How the request was routed (provider, fallback, cache, latency).",
+      "properties": {
+        "cache_hit": {
+          "default": false,
+          "description": "True if the result was served from cache.",
+          "title": "Cache Hit",
+          "type": "boolean"
+        },
+        "fallback_used": {
+          "default": false,
+          "description": "True if the primary provider failed and a fallback served the result.",
+          "title": "Fallback Used",
+          "type": "boolean"
+        },
+        "latency_ms": {
+          "default": 0,
+          "description": "End-to-end routing latency in milliseconds.",
+          "title": "Latency Ms",
+          "type": "integer"
+        },
+        "provider": {
+          "description": "Name of the upstream provider that served the result.",
+          "title": "Provider",
+          "type": "string"
+        },
+        "providers_tried": {
+          "description": "Providers attempted, in order.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Providers Tried",
+          "type": "array"
+        }
+      },
+      "required": [
+        "provider"
+      ],
+      "title": "RoutingMeta",
+      "type": "object"
+    },
+    "title": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "default": null,
+      "title": "Title"
+    },
+    "token": {
+      "default": "USDC",
+      "description": "Token actually used for payment (USDC, USDT or FREE).",
+      "title": "Token",
+      "type": "string"
+    },
+    "truncated": {
+      "default": false,
+      "title": "Truncated",
+      "type": "boolean"
+    },
+    "url": {
+      "title": "Url",
+      "type": "string"
+    },
+    "word_count": {
+      "title": "Word Count",
+      "type": "integer"
+    },
+    "x_payment_response": {
+      "description": "Base64-encoded x402 settlement receipt (mirrors the X-PAYMENT-RESPONSE header). Only present when the call was paid with `x_payment`.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "freshness_seconds",
+    "cost_usdc",
+    "routing",
+    "url",
+    "final_url",
+    "markdown",
+    "word_count",
+    "char_count",
+    "extraction_method",
+    "include_links",
+    "include_images"
+  ],
+  "type": "object"
+}

Changed3 schema fields changedSep 13, 2026

addedInput schema / properties / include_images / default
```
Added value: +false
```
addedInput schema / properties / include_links / default
```
Added value: +true
```

changedInput schema / properties / wallet / description

Previous value: -"Your EVM wallet address (0x...). Unlocks the free tier (3 calls per wallet)."New value: +"Your EVM wallet address (0x...). Unlocks the free tier (3 free calls per wallet)."

AddedSep 13, 2026

TDQS

A4.5/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description goes well beyond the readOnlyHint and openWorldHint annotations by disclosing the extraction pipeline (Trafilatura with markdownify fallback), the filtering behavior (boilerplate/nav/ads removed), SSRF guarding, the 3 MB limit, and the payment/free-tier model. This gives the agent strong expectations about side effects, constraints, and costs.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three dense, front-loaded sentences: purpose and output, extraction behavior and constraints, then payment terms. Every sentence adds distinct and necessary information with no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has an output schema and six parameters, the description covers the essential operational context: input URL type, output format, content filtering, fallback behavior, size cap, payment requirements, and free-tier usage. Nothing critical for invoking the tool correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all six parameters are already documented in the input schema. The description adds high-level context about optional links/images and payment, but it does not materially go beyond what the schema already states for any parameter.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb and resource ('Convert any public web page into clean, LLM-ready Markdown'), then elaborates exactly what is preserved and removed. This clearly distinguishes the tool from siblings like web_search or pdf_text by its output format and extraction-only scope.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly says when to use the tool: when you have a public web page URL and want clean Markdown. It also states exclusions such as 'public' pages and a 3 MB HTML limit, but it does not explicitly name alternatives or say when to prefer extract_clean or web_crawl instead.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Agent Margin Router tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.