Firecrawl MCP Server

Overview Schema Source Related Servers Discussions Admin

Firecrawl file parsing

firecrawl_parse

Read-only

Parse local HTML, PDF, Word, RTF, OpenDocument, and spreadsheet files into Markdown, HTML, links, summaries, targeted answers, or schema-based JSON for AI workflows.

Instructions

Parse one supported document into markdown, HTML, links, summary, targeted answers, or JSON matching a schema. Supported inputs include common HTML, PDF, Word, RTF, OpenDocument, and spreadsheet files; PDF parsing can be bounded with pdfOptions.maxPages.

Local MCP reads filePath from the server filesystem. Hosted MCP uses two calls: first provide filePath to receive upload instructions, upload locally, then call again with the returned uploadRef; do not send both fields together. Remote web URLs belong in firecrawl_scrape.

Set redactPII to request redaction of personally identifiable information in the returned content. zeroDataRetention requires an eligible authenticated account; omit it for anonymous keyless use. Returns upload instructions for hosted phase one or parsed document content for the final call. Authenticated final responses can include a data.metadata.scrapeId for optional parse feedback.

Input Schema

Name	Required	Description
`proxy`	No
`maxAge`	No	Ignored: parse never reuses or stores indexed content.
`formats`	No
`parsers`	No
`filePath`	Yes	Absolute or relative path to a local file to parse. Supported: .html, .htm, .pdf, .docx, .doc, .odt, .rtf, .xlsx, .xls
`redactPII`	No
`pdfOptions`	No
`contentType`	No	Optional MIME type override. If omitted, the server infers the file kind from the extension.
`excludeTags`	No
`includeTags`	No
`jsonOptions`	No
`queryOptions`	No
`storeInCache`	No
`onlyMainContent`	No
`zeroDataRetention`	No
`removeBase64Images`	No
`skipTlsVerification`	No

Output Schema

Name	Required	Description
`raw`	No	Response body when it was not JSON.
`data`	No	Parsed document content; can include `data.metadata.scrapeId` for parse feedback.
`mode`	No	Which phase of the hosted flow produced this response.
`error`	No	Error message or error object when the call did not succeed.
`notes`	No	Hosted phase one: constraints on completing the upload flow.
`upload`	No	Hosted phase one: how to upload the local file.
`message`	No	Guidance for the next call.
`success`	No	Whether the API call succeeded.
`warning`	No	Non-fatal warning about the result.
`nextToolCall`	No	Hosted phase one: the second `firecrawl_parse` call to make once the upload succeeds, as `{name, arguments}`.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv3.36.1Nov 5, 2026
- changedInput schema / properties / jsonOptions / properties / schema / additionalProperties
```
Previous value: -{}New value: +true
```

Changed4 schema fields changedv1.0.2Oct 29, 2026

changedInput schema / $schema

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

addedInput schema / properties / maxAge / description

Added value: +"Ignored: parse never reuses or stores indexed content."

changedInput schema / properties / queryOptions / required

Previous value: -[
-  "prompt",
-  "mode"
-]New value: +[
+  "prompt"
+]

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "description": "Parsed document content, or the upload instructions for the hosted two-call flow.",
+  "properties": {
+    "data": {
+      "description": "Parsed document content; can include `data.metadata.scrapeId` for parse feedback."
+    },
+    "error": {
+      "description": "Error message or error object when the call did not succeed."
+    },
+    "message": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Guidance for the next call."
+    },
+    "mode": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Which phase of the hosted flow produced this response."
+    },
+    "nextToolCall": {
+      "description": "Hosted phase one: the second `firecrawl_parse` call to make once the upload succeeds, as `{name, arguments}`."
+    },
+    "notes": {
+      "anyOf": [
+        {
+          "items": {
+            "type": "string"
+          },
+          "type": "array"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Hosted phase one: constraints on completing the upload flow."
+    },
+    "raw": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Response body when it was not JSON."
+    },
+    "success": {
+      "anyOf": [
+        {
+          "type": "boolean"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Whether the API call succeeded."
+    },
+    "upload": {
+      "additionalProperties": false,
+      "description": "Hosted phase one: how to upload the local file.",
+      "properties": {
+        "command": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "Local command that performs the upload. It carries no Firecrawl API key."
+        },
+        "expiresAt": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "When the upload URL expires."
+        },
+        "fields": {
+          "description": "Form fields the upload request must send, for a POST upload."
+        },
+        "headers": {
+          "description": "Headers the upload request must send."
+        },
+        "maxSizeBytes": {
+          "anyOf": [
+            {
+              "type": "number"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "Largest file the upload URL accepts."
+        },
+        "method": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "HTTP method for the upload."
+        },
+        "uploadRef": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "Reference to pass back on the second call."
+        },
+        "uploadUrl": {
+          "anyOf": [
+            {
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ],
+          "description": "URL to upload the local file to."
+        }
+      },
+      "type": "object"
+    },
+    "warning": {
+      "anyOf": [
+        {
+          "type": "string"
+        },
+        {
+          "type": "null"
+        }
+      ],
+      "description": "Non-fatal warning about the result."
+    }
+  },
+  "type": "object"
+}

First observedv1.0.1Jul 5, 2026

TDQS

A4.3/5.0

Behavior5/5

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

Beyond the annotations (which only declare read-only, non-destructive, closed-world), it discloses the hosted two-phase protocol, that phase one returns upload instructions rather than content, that `zeroDataRetention` requires an eligible authenticated account and must be omitted for keyless use, PII redaction behavior, and the optional `data.metadata.scrapeId` for feedback. That is substantial context an agent could not infer from annotations alone.

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 the core action and output modes, then organized into input types, local-vs-hosted protocol, and option notes. Dense but each sentence carries operational content; the hosted-flow paragraph is the longest and could be tightened, but nothing is filler.

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

Completeness3/5

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

Return values are covered by the output schema and the description usefully notes phase-one vs final responses, so that burden is met. However, for a 17-parameter tool with nested objects, the vast majority of options are unexplained, and the `uploadRef` instruction is not representable in the published schema — a gap an agent will hit on the hosted path.

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

Parameters2/5

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

With 17 parameters and only 18% schema description coverage, the description must carry most of the semantic load, yet it explains only `filePath`, `pdfOptions.maxPages`, `redactPII`, and `zeroDataRetention`. Formats, `parsers`, `jsonOptions`/`queryOptions`, `includeTags`/`excludeTags`, `onlyMainContent`, `storeInCache`, `proxy`, `maxAge`, and `contentType` are left undefined. Worse, the description instructs sending `uploadRef`, which does not exist in the schema (which is `additionalProperties: false` and requires `filePath`), creating ambiguity about how the second hosted call is actually formed.

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 ('Parse one supported document') plus the concrete output modes (markdown, HTML, links, summary, targeted answers, schema JSON) and the accepted input families. It also explicitly separates itself from `firecrawl_scrape` for remote web URLs, so an agent can route without opening either schema.

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 explicit when/when-not: local MCP reads `filePath` from the server filesystem, hosted MCP requires two calls (filePath → upload → uploadRef), never both fields together, and remote URLs belong to `firecrawl_scrape`. This is a genuine decision procedure, not implied usage.

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