FileForge

Overview Schema Source Related Servers Discussions Admin

fileforge_scrape_action

Extract structured content from pages, tabs, or sites—text, links, tables, metadata, assets, and design tokens—to store, query, or compare results for analysis.

Instructions

Extract structured content from pages, tabs or a whole site: text, links, tables, metadata, assets and design tokens. Prefer query_scrape; the others are older single-purpose actions. It can't click or submit; use fileforge_page_interaction for that. crawl_site opens real tabs in your signed-in session, up to 50 pages by default. Query stored links and tables with fileforge_sql_action. Needs the extension.

Input Schema

Name	Required	Description	Default
`url`	No	Page URL. Omit to use the active tab where the action allows it.
`urls`	No	URLs to process.
`query`	No	Sections to return and their options. Takes priority over preset. Sections: page, metadata, headings, sections, links, images, assets, tables, jsonLd, article, contacts, pricing, products, contentQuality, warnings, summary, snapshot.
`store`	No	Save the result in ~/.fileforge and return a ref.
`tabId`	No	Tab ID, or "active" for the active tab.	active
`action`	Yes	What to do. Required parameters are in brackets. query_scrape: GraphQL-style projection over page content (presets + custom queries). compare_scrape_snapshots (beforeSnapshotId, afterSnapshotId): Diff two stored scrape snapshots by ID. scrape_page: Scrape sections of a single page. extract_assets: Extract assets (images, videos, files) from a page. extract_tables: Extract tables from the active or specified page. extract_contacts: Extract email/phone/address contact data. extract_metadata: Extract page metadata (title, OG, JSON-LD). extract_design_tokens: Extract colors, fonts, and design tokens. extract_all_debug: Extract everything (debug only — large payload). scrape_tabs (tabIds): Scrape multiple tabs. scrape_urls (urls): Open and scrape a list of URLs. crawl_site (url): Crawl a site up to maxPages. research_site (url): Research a site (multi-step crawl + extract). analyze_site (url): Analyze a site. extract_resource (url): Fetch a specific page resource.
`preset`	No	Named set of query modules.
`source`	No	What to scrape: { url } or { tabId }. Defaults to the active tab.
`tabIds`	No	Tab IDs to act on.
`project`	No	Project name that groups stored results.	default
`maxPages`	No	Maximum pages to visit.
`sections`	No	Page sections to scrape. Defaults to all of them.
`afterSnapshotId`	No	ID of the later snapshot.
`beforeSnapshotId`	No	ID of the earlier snapshot.

Output Schema

Name	Required	Description
`ok`	Yes	False if the action failed or a check did not pass.
`ref`	No	Stored result ID, for fileforge_result_action.
`data`	No	The result. Its shape depends on the action.
`meta`	No	Tool, action and context.
`error`	No	Failure code and message.
`warnings`	No	Notes about a call that still ran.

Schema Changelog

Changes observed during successful MCP inspections.

Changed17 schema fields changedv0.3.0Nov 5, 2026

changedInput schema / properties / action / description

Previous value: -"The scrape action to perform."New value: +"What to do. Required parameters are in brackets. query_scrape: GraphQL-style projection over page content (presets + custom queries). compare_scrape_snapshots (beforeSnapshotId, afterSnapshotId): Diff two stored scrape snapshots by ID. scrape_page: Scrape sections of a single page. extract_assets: Extract assets (images, videos, files) from a page. extract_tables: Extract tables from the active or specified page. extract_contacts: Extract email/phone/address contact data. extract_metadata: Extract page metadata (title, OG, JSON-LD). extract_design_tokens: Extract colors, fonts, and design tokens. extract_all_debug: Extract everything (debug only — large payload). scrape_tabs (tabIds): Scrape multiple tabs. scrape_urls (urls): Open and scrape a list of URLs. crawl_site (url): Crawl a site up to maxPages. research_site (url): Research a site (multi-step crawl + extract). analyze_site (url): Analyze a site. extract_resource (url): Fetch a specific page resource."

addedInput schema / properties / afterSnapshotId / description
```
Added value: +"ID of the later snapshot."
```
addedInput schema / properties / beforeSnapshotId / description
```
Added value: +"ID of the earlier snapshot."
```
addedInput schema / properties / maxPages / description
```
Added value: +"Maximum pages to visit."
```
addedInput schema / properties / preset / description
```
Added value: +"Named set of query modules."
```

addedInput schema / properties / preset / enum

Added value: +[
+  "basic",
+  "page",
+  "links",
+  "tables",
+  "seo_audit",
+  "landing_page_audit",
+  "article_extract",
+  "research_extract",
+  "competitor_snapshot",
+  "pricing_audit",
+  "structured_data",
+  "contact_discovery",
+  "asset_audit"
+]

addedInput schema / properties / project / description

Added value: +"Project name that groups stored results."

addedInput schema / properties / query / description

Added value: +"Sections to return and their options. Takes priority over preset. Sections: page, metadata, headings, sections, links, images, assets, tables, jsonLd, article, contacts, pricing, products, contentQuality, warnings, summary, snapshot."

addedInput schema / properties / sections / description

Added value: +"Page sections to scrape. Defaults to all of them."

addedInput schema / properties / sections / items / enum

Added value: +[
+  "files",
+  "images",
+  "links",
+  "pageLinks",
+  "emails",
+  "phones",
+  "tables",
+  "videos",
+  "colors",
+  "addresses",
+  "locations",
+  "fonts",
+  "lists",
+  "icons",
+  "qrCodes",
+  "structuredData",
+  "network",
+  "metadata",
+  "assets",
+  "contacts",
+  "designTokens",
+  "dataUris"
+]

addedInput schema / properties / source / description

Added value: +"What to scrape: { url } or { tabId }. Defaults to the active tab."

addedInput schema / properties / store / description

Added value: +"Save the result in ~/.fileforge and return a ref."

addedInput schema / properties / tabId / description

Added value: +"Tab ID, or \"active\" for the active tab."

addedInput schema / properties / tabIds / description
```
Added value: +"Tab IDs to act on."
```

addedInput schema / properties / url / description

Added value: +"Page URL. Omit to use the active tab where the action allows it."

addedInput schema / properties / urls / description
```
Added value: +"URLs to process."
```

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "description": "query_* actions return their modules as extra top-level fields.",
+  "properties": {
+    "data": {
+      "description": "The result. Its shape depends on the action."
+    },
+    "error": {
+      "description": "Failure code and message."
+    },
+    "meta": {
+      "description": "Tool, action and context."
+    },
+    "ok": {
+      "description": "False if the action failed or a check did not pass.",
+      "type": "boolean"
+    },
+    "ref": {
+      "description": "Stored result ID, for fileforge_result_action."
+    },
+    "warnings": {
+      "description": "Notes about a call that still ran."
+    }
+  },
+  "required": [
+    "ok"
+  ],
+  "type": "object"
+}

First observedv0.2.1Nov 5, 2026

TDQS

A4.1/5.0

Behavior4/5

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

Annotations only declare the generic profile (openWorld, not read-only, not destructive); the description adds real behavioral context beyond them: crawl_site 'opens real tabs in your signed-in session, up to 50 pages by default', which is a meaningful side effect on the user's browser session, and 'Needs the extension', a hard prerequisite. It does not mention persistence to ~/.fileforge or payload size for extract_all_debug, so it is strong but not exhaustive.

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-loaded with purpose, then alternatives, then constraints — a sensible ordering with no filler sentences. It is telegraphic in places ('Needs the extension.'), which is efficient but leaves the reader to infer the consequence of a missing extension.

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

Completeness4/5

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

For a 14-parameter dispatcher over 15 actions with nested objects, the description covers the essentials an agent needs before calling: prerequisites, side effects, and where to go for adjacent capabilities, while the schema and output schema carry parameter and return detail. What is missing is a brief signal about which actions require which input shape (url vs urls vs tabIds) at the top level.

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% and every parameter, including the 15-value action enum and the presets, is already documented inline, so the baseline is 3. The description adds only one genuine parameter fact beyond the schema — the default page cap for crawl_site — and otherwise repeats enum semantics rather than clarifying how url/urls/source/tabId interact when several are supplied.

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?

Opens with a specific verb and resource set: 'Extract structured content from pages, tabs or a whole site' followed by the content types returned (text, links, tables, metadata, assets, design tokens). It also distinguishes itself from siblings by name (page_interaction for clicking, sql_action for stored data), so an agent can place it in the toolset without opening the schema.

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?

Explicit routing guidance: 'Prefer query_scrape; the others are older single-purpose actions', plus hard exclusions ('can't click or submit; use fileforge_page_interaction') and a handoff ('Query stored links and tables with fileforge_sql_action'). The gap is that the many site-level actions (crawl_site vs research_site vs analyze_site) are not disambiguated from each other, leaving the agent to guess among overlapping crawl variants.

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