Talonic

talonicdev/talonic-mcp

Talonic wraps document intelligence into eight MCP tools that turn PDFs, scans, images, and spreadsheets into schema-validated JSON. You get talonic_extract for structured data extraction with confidence scores, talonic_to_markdown for OCR output, talonic_search and talonic_filter for querying your workspace, and schema management tools for saving and reusing extraction patterns. Supports stdio transports for Claude Desktop, Cursor, and VS Code extensions, plus a hosted MCP endpoint for Claude.ai with OAuth or API key auth. Free tier gives you 50 extractions per day. Reach for this when you need reliable table parsing, form field extraction, or invoice line items instead of fighting raw OCR output and prompt engineering your way to consistency.

GitHub stars: 5
tools: 8
auth: Secret needed
transport: STDIO
env var: 1
last updated: Oct 29, 2026

View on GitHub

Tools

8 tools

Public tool metadata for what this MCP can expose to an agent.

talonic_list_schemasSTATUS: stable. List all saved schemas in the user's Talonic workspace. Returns each schema with its id (UUID), short_id (SCH-XXXXXXXX), name, description, version, field count, and full JSON Schema definition. Either id form is accepted by talonic_extract's `schema_id` parame...

STATUS: stable. List all saved schemas in the user's Talonic workspace. Returns each schema with its id (UUID), short_id (SCH-XXXXXXXX), name, description, version, field count, and full JSON Schema definition. Either id form is accepted by talonic_extract's `schema_id` parame...

No parameter schema in public metadata yet.

talonic_save_schemaSTATUS: stable. Save a schema definition to the user's Talonic workspace so it can be reused across future extractions. Returns the saved schema with its newly assigned id and short_id. USE WHEN: - The user asks to save a schema, store a template, or reuse the schema across do...3 params

STATUS: stable. Save a schema definition to the user's Talonic workspace so it can be reused across future extractions. Returns the saved schema with its newly assigned id and short_id. USE WHEN: - The user asks to save a schema, store a template, or reuse the schema across do...

Parameters* required

definitionobject

Schema definition. Most reliable: full JSON Schema {type:'object', properties:{...}}. Also accepted: a flat key-type map {field_name:'string', amount:'number'} which the API normalises.

descriptionstring

Optional description of what this schema extracts and when to use it.

namestring

Human-readable name for the schema, e.g. 'Standard Invoice'.

talonic_get_documentSTATUS: stable. Fetch full metadata for a single document already in the user's Talonic workspace. Returns id, filename, page count, detected document type, language, processing log, and link URLs (self, extractions, dashboard). USE WHEN: - You need details about a specific do...1 params

STATUS: stable. Fetch full metadata for a single document already in the user's Talonic workspace. Returns id, filename, page count, detected document type, language, processing log, and link URLs (self, extractions, dashboard). USE WHEN: - You need details about a specific do...

Parameters* required

document_idstring

The Talonic document ID. Get this from a previous talonic_extract or talonic_search response.

talonic_searchSTATUS: stable. Search the user's Talonic workspace for documents, fields, sources, or schemas matching a query. Returns ranked results across all entity types in one call. USE WHEN: - The user wants to find documents but does not know the exact filename or id. - The query is...2 params

STATUS: stable. Search the user's Talonic workspace for documents, fields, sources, or schemas matching a query. Returns ranked results across all entity types in one call. USE WHEN: - The user wants to find documents but does not know the exact filename or id. - The query is...

Parameters* required

limitinteger

Maximum results per entity type. Default: 5. Increase for broader exploration.

querystring

Natural-language search query, e.g. 'indemnification clauses' or 'Acme invoices Q4'.

talonic_filterSTATUS: stable. Field-name resolution is server-side. The `is_not_empty` operator is intentionally not exposed in v0.1; see workaround below. Filter the user's Talonic documents by extracted field values using composable conditions. Conditions accept either a canonical field n...6 params

STATUS: stable. Field-name resolution is server-side. The `is_not_empty` operator is intentionally not exposed in v0.1; see workaround below. Filter the user's Talonic documents by extracted field values using composable conditions. Conditions accept either a canonical field n...

Parameters* required

conditionsarray

One or more filter conditions, AND-ed together.

limitinteger

Results per page. Default 50 server-side.

pageinteger

Page number for pagination.

searchstring

Optional free-text search applied alongside the filters.

sortobject

Optional sort by a field.

source_connection_idstring

Optionally scope to a specific source connection.

talonic_to_markdownSTATUS: stable. Get the OCR-converted markdown for a document. Accepts an existing document_id, raw file bytes (base64), a local file path, or a URL. When given a raw file, the tool ingests it via extract first and then returns the markdown. USE WHEN: - The user wants the full...5 params

STATUS: stable. Get the OCR-converted markdown for a document. Accepts an existing document_id, raw file bytes (base64), a local file path, or a URL. When given a raw file, the tool ingests it via extract first and then returns the markdown. USE WHEN: - The user wants the full...

Parameters* required

document_idstring

The Talonic document id whose markdown you want. Get this from a previous talonic_extract or talonic_search response.

file_datastring

Base64-encoded file bytes. Recommended path when the agent already has the file in memory (e.g., the user attached a PDF to the conversation). Pair with `filename` so MIME type can be inferred.

file_pathstring

Local path to a document file. Only works if the MCP server has read access to that path. In sandboxed chat clients (Claude Desktop, Cowork) use `file_data` instead.

file_urlstring

URL to a document file. The Talonic API fetches it server-side.

filenamestring

Original filename including extension, e.g. 'invoice.pdf'. Used to infer MIME type when uploading via `file_data`. Required when `file_data` is provided.

talonic_extractSTATUS: stable. Production-safe when called with a schema. Schema-less extraction is disabled at the MCP layer. Extract structured, schema-validated data from a document using Talonic. Returns clean JSON matching the schema, with per-field confidence scores and metadata about...10 params

STATUS: stable. Production-safe when called with a schema. Schema-less extraction is disabled at the MCP layer. Extract structured, schema-validated data from a document using Talonic. Returns clean JSON matching the schema, with per-field confidence scores and metadata about...

Parameters* required

document_idstring

ID of a document already in the workspace, to re-extract with a new schema.

file_datastring

Base64-encoded file bytes. Recommended path when the agent already has the file in memory (e.g., the user attached a PDF to the conversation). Pair with `filename` so MIME type can be inferred. Wor...

file_pathstring

Local path to a document file. Only works if the MCP server has read access to that path. In sandboxed chat clients (Claude Desktop, Cowork) where uploads land in a host-owned directory, use `file_...

file_urlstring

URL to a document file. The Talonic API fetches it server-side. Use this for documents already on the public web.

filenamestring

Original filename including extension, e.g. 'invoice.pdf'. Used to infer MIME type when uploading via `file_data`. Required when `file_data` is provided.

include_markdownboolean

Include OCR-converted markdown in the response alongside structured data.

include_provenanceboolean

Include per-field provenance (source_text, section, page) showing where each value was found in the document.

instructionsstring

Natural-language guidance for the extractor, e.g. 'Focus on the billing section. Amounts are in EUR.'

schemaobject

Inline schema definition. REQUIRED unless `schema_id` is provided. Recommended: full JSON Schema {type:'object', properties:{...}}. Also accepted: flat key-type map {field_name:'string', amount:'nu...

schema_idstring

ID of a saved schema. REQUIRED unless `schema` is provided. Accepts UUID or SCH-XXXXXXXX short id from talonic_list_schemas. Mutually exclusive with `schema`.

talonic_get_balanceSTATUS: stable. Read the user's current Talonic credit balance, EUR value, 30-day burn rate, projected runway, tier, and next-tier-reset timestamp. Use this to make budget- aware decisions before kicking off large batches or re-extractions. USE WHEN: - The user asks how many c...

STATUS: stable. Read the user's current Talonic credit balance, EUR value, 30-day burn rate, projected runway, tier, and next-tier-reset timestamp. Use this to make budget- aware decisions before kicking off large batches or re-extractions. USE WHEN: - The user asks how many c...

No parameter schema in public metadata yet.

Configuration

TALONIC_API_KEY*secret

Your Talonic API key. Get one at https://app.talonic.com.

Readme

@talonic/mcp

Official Talonic MCP server. Give any AI agent the ability to extract structured, schema-validated data from any document — PDFs, scans, invoices, contracts, forms — via the Model Context Protocol.

View the full README on GitHub

Details

Repository: github.com/talonicdev/talonic-mcp
Website: talonic.com
Categories: Search & Web Crawling, Data & Analytics
Package: @talonic/mcp
Transport: STDIO
Auth: Required
Registry: MCP registry, active

Related Search & Web Crawling MCP servers

View all

Biorxiv Mcp Servercyanheads/biorxiv-mcp-serverSearch and retrieve bioRxiv and medRxiv preprints — by DOI, date interval, or keyword — via MCP.4stars

Socrata Mcp Servercyanheads/socrata-mcp-serverSearch and query government open-data portals (Socrata SODA API).4stars

NewzAI MCP Servergauraviitkgp/news-mcpNews MCP: real-time headlines & custom news search across 7 regions. Free, just sign in with Google.4stars

JVM Source Lenssintexer/jvm-source-lensGradle-accurate JVM classpath; fetch Java source, signatures, and class structure for agents.4stars

Apple Voice Memo Mcpjwulff/apple-voice-memo-mcpAccess Apple Voice Memos on macOS. List, get audio, extract and generate transcripts.4stars

Data Aggregatormusharna/data-aggregator-mcpFind & fetch research datasets across 21 archives, omics registries, and literature sources.4stars