CrawlForge MCP Server

agent

Read-only

Research any question across the web without providing URLs: plan searches, fetch and filter pages, then return a prose or structured answer.

Instructions

Use this when you need an autonomous agent to research, navigate, and synthesise an answer from the web - no URLs required. The agent plans search queries, fetches and filters relevant pages, and returns a prose or structured answer. model:"pro" uses deep multi-source research. Hard limits: maxSteps<=10, maxUrls<=20, 120s wall-clock. Confirms before pro runs. Degraded-but-useful output if no LLM keys/Ollama. Not for a URL you already have (scrape) or a question one search answers (search_web). Pages that block a plain fetch are retried in the stealth browser automatically (at most 2 a run, URLs you name first; evidence marked via:"stealth"). Cost: 18 credits at most - 8, plus 5 per stealth retry that gets the page; a retry that is blocked again is free. Example: agent({prompt:"What are the top 5 MCP servers in 2025?", maxUrls:10})

Input Schema

TableJSON Schema

Name	Required	Description	Default
`urls`	No	Optional seed URLs to include (max 20)
`model`	No	"default" = SamplingClient loop (no keys needed); "pro" = full ResearchOrchestrator	default
`prompt`	Yes	Natural-language task or question
`schema`	No	Optional JSON schema for structured output
`maxUrls`	No	Max URLs to fetch (hard cap: 20)
`maxSteps`	No	Max fetch iterations (hard cap: 10)
`max_inline_chars`	No	Largest result to return inline, in characters of its JSON. Over it, the call returns a preview plus a result_handle for read_result instead of the whole result (default 40,000; env CRAWLFORGE_MAX_INLINE_CHARS)

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv6.19.2Oct 6, 2026

addedInput schema / properties / max_inline_chars

Added value: +{
+  "description": "Largest result to return inline, in characters of its JSON. Over it, the call returns a preview plus a result_handle for read_result instead of the whole result (default 40,000; env CRAWLFORGE_MAX_INLINE_CHARS)",
+  "maximum": 10000000,
+  "minimum": 1000,
+  "type": "integer"
+}

Changed2 schema fields changedv6.0.0Sep 6, 2026
- removedInput schema / additionalProperties
```
Removed value: -false
```
- addedInput schema / properties / schema / propertyNames
```
Added value: +{
+  "type": "string"
+}
```

Changed1 schema field changedv5.0.4Aug 21, 2026

changedInput schema / $schema

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

First observedv4.10.0Jul 29, 2026

TDQS

A4.9/5.0

Behavior5/5

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

Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint. The description adds substantial non-obvious behavior: maxSteps<=10, maxUrls<=20, 120s wall-clock, pro confirmation, degraded output without LLM keys/Ollama, stealth retry rules, evidence marking, and credit cost. This goes well beyond structured annotations.

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?

Front-loads the core purpose, then alternates and exclusions, then hard limits, degradation behavior, stealth behavior, cost, and an example. Despite covering many operational details, every sentence carries actionable information and there is 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?

For a complex autonomous research tool with no output schema but rich annotations and 100% schema coverage, the description supplies routing rules, runtime limits, cost model, fallback behavior, structured-output option, and a usage example. An agent has enough context to select and invoke it correctly.

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, but the description adds real semantic detail beyond the schema: 'model:"pro" uses deep multi-source research,' the hard limits for maxSteps and maxUrls, the 120s wall-clock constraint, and a concrete call example. It does not add much beyond the schema for prompt/schema/urls individually, but the model and limit context is useful.

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 precise verb-and-resource purpose: an autonomous agent that researches, navigates, and synthesises an answer from the web without requiring URLs. It explicitly distinguishes itself from sibling tools by ruling out 'a URL you already have (scrape)' and 'a question one search answers (search_web)'.

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?

Provides explicit when-to-use guidance ('Use this when you need an autonomous agent...') and when-not-to-use alternatives, naming scrape and search_web directly. Also clarifies model:'pro' behavior, confirmation behavior, and cost/limit conditions that affect invocation choice.

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

agent

Instructions

Input Schema

Schema Changelog

TDQS

Other Tools