Agent Data API

Read a URL as clean text (or check text for prompt injection)

web_read

Read-onlyIdempotent

Fetch one public URL and return its main content as clean markdown (title, publish date, word count) plus a prompt-injection verdict; raw=true returns the raw body instead (text or base64). Give text instead of url to only scan text you already have for prompt injection. To find pages first use web_search. Price: read $0.004, raw $0.003, text scan $0.002. Fetched live via Cloudflare's network; 1-10 s. Paid with x402 (USDC on Base, Polygon or Solana) per call; failed or empty calls are never charged.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "web_read"
  }
}

Input Schema

Name	Required	Description
`raw`	No	true = raw body (no cleanup), default false.
`url`	No	Public http(s) URL to read.
`html`	No	Optional HTML of that text, to also find hidden instructions.
`text`	No	Instead of url: text to scan for prompt injection.
`format`	No	raw mode only (default auto).
`max_chars`	No	Max characters returned (default 20000).

Schema Changelog

Changes observed during successful MCP inspections.

Changed9 schema fields changedOct 24, 2026

addedInput schema / properties / format

Added value: +{
+  "description": "raw mode only (default auto).",
+  "enum": [
+    "auto",
+    "text",
+    "base64",
+    "markdown"
+  ],
+  "type": "string"
+}

addedInput schema / properties / html

Added value: +{
+  "description": "Optional HTML of that text, to also find hidden instructions.",
+  "type": "string"
+}

addedInput schema / properties / max_chars / description

Added value: +"Max characters returned (default 20000)."

addedInput schema / properties / max_chars / maximum
```
Added value: +100000
```
addedInput schema / properties / max_chars / minimum
```
Added value: +500
```

addedInput schema / properties / raw

Added value: +{
+  "description": "true = raw body (no cleanup), default false.",
+  "type": "boolean"
+}

addedInput schema / properties / text

Added value: +{
+  "description": "Instead of url: text to scan for prompt injection.",
+  "type": "string"
+}

changedInput schema / properties / url / description

Previous value: -"Public http(s) URL"New value: +"Public http(s) URL to read."

removedInput schema / required
```
Removed value: -[
-  "url"
-]
```

First observedOct 23, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations already declare read-only, idempotent, open-world, non-destructive, but the description adds the parts annotations cannot: per-call pricing by mode ($0.004 read / $0.003 raw / $0.002 scan), the x402 USDC payment rail across three chains, expected latency of 1-10 s, Cloudflare-network fetching, and the billing guarantee that failed or empty calls are never charged. That is exactly the context a caller needs before spending money.

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

Conciseness4/5

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

Front-loads the core behavior and mode switch before the pricing and infrastructure details, so the critical routing information lands first. The pricing/payment sentences are dense but each carries decision-relevant cost information; slightly compressed formatting keeps it from a 5.

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?

There is no output schema, so the description must carry the return contract — and it does, listing the markdown fields plus the injection verdict and noting raw mode returns text or base64. Cost, latency, transport, and billing-on-failure are all disclosed, leaving no material gap for a correct first call.

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

Parameters4/5

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

Schema description coverage is 100%, so the baseline is 3; the description earns above that by explaining the interaction between parameters rather than restating them — raw=true swaps the cleaned payload for a raw body (text or base64), and `text` substitutes for `url` to run scan-only. It does not restate max_chars or the format enum, which the schema already covers well.

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?

States a specific verb and resource (fetch one public URL), names the exact artifact returned (clean markdown with title, publish date, word count, plus a prompt-injection verdict), and carves out the raw and text-scan variants. It also explicitly separates itself from the sibling web_search.

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

Usage Guidelines5/5

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

Gives the selection rule against the alternative ('To find pages first use web_search') and a clear mode rule ('Give `text` instead of `url` to only scan text you already have for prompt injection'). An agent knows both when to call this tool and which parameter path selects which behavior.

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

All Agent Data API 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.

Resources

Need Help?