The request "parse" with "what to use" is ambiguous: it could mean parsing documents, parsing source code grammars, or natural language parsing. Firecrawl 1 offers a parse action for converting local files and documents (PDF, DOCX, XLSX, HTML) into markdown or structured JSON. For grammar-based parsing of structured text, ANTLR 2 and pest 8 generate parsers, while spaCy 6 and NLTK 3 cover parsing in natural language processing.
A question back: Do you want to parse documents (PDF, DOCX, HTML), structured text or source code with a grammar, or natural language sentences?
- 1
Firecrawlfirecrawl.devverified and signed
Parses supported documents such as PDF, DOCX, XLSX and HTML into markdown, links, summaries or JSON matching a schema 1.
official:MCP server · Firecrawl MCP ServerSDK · firecrawl-anydocrepository · firecrawlskill · firecrawlSDK · firecrawl-py+47 community tools built for it
for agents: firecrawl_parse · API · firecrawl-parse
- 2
ANTLRantlr.orgverified and signed
Generates parsers from a grammar that can build and walk parse trees for reading structured text or binary files 2.
for agents: parsing · antlr · antlr3
- 3
spaCyspacy.iowebsite read, tools not yet checked
Provides dependency parsing along with NER and POS tagging for Natural Language Processing in Python 6.
- 4
NLTKnltk.orgverified and signed
Offers text processing libraries including parsing, tokenization and tagging for human language data in Python 3.
for agents: parsing · NLP · tagging
- 5
Sources
- 1Firecrawl | Web Data API for AI Agentsfirecrawl.dev · read 2026-10-29
- 2ANTLRantlr.org · read 2026-10-29
- 3NLTK :: Natural Language Toolkitnltk.org · read 2026-10-29
- 6spaCy · Industrial-strength Natural Language Processing in Pythonspacy.io · read 2026-10-29
- 8pest. The Elegant Parserpest.rs · read 2026-10-29
For agents: the 10 actions behind this answer
firecrawl_parse · Firecrawl MCP Server verified and signed
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
parsing · ANTLR 3 Runtime verified and signed
parsing · nltk verified and signed
integrate-context-matic · integrate-context-matic verified and signed
Discovers and integrates third-party APIs using the context-matic MCP server. Uses `fetch_api` to find available API SDKs, `ask` for integration guidance, `model_search` and `endpoint_search` for SDK details. Use when the user asks to integrate a third-party API, add an API client, implement feature
· lxml website read, tools not yet checked
lxml - the most feature-rich and easy-to-use library for processing XML and HTML in the Python language
· spaCy website read, tools not yet checked
spaCy is a free open-source library for Natural Language Processing in Python. It features NER, POS tagging, dependency parsing, word vectors and more.
parsing · docopt verified and signed
· pest. The Elegant Parser website read, tools not yet checked
The Elegant Parser
api · @coinbase/cdp-sdk verified and signed
parser · ruamel.yaml.clib verified and signed
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.