Connect MESHH to your AI workspace.

Use MCP to bring MESHH's Japanese public-source research layer into Claude, ChatGPT, Codex, or internal tools.

MCP endpoint

https://meshh.ai/mcp

Anthropic

Claude

Connect MESHH into Claude's MCP flow and use Japanese public-source research directly inside conversations.

OpenAI

ChatGPT

Open the ChatGPT-specific MCP setup flow for MESHH and attach the same research layer to your workspace.

MCP Tools

Research actions available after connection

get_stats

Inspect corpus coverage, counts by type, and language/date range metadata.

list_items

Browse corpus items by source, subtype, language, and date without running text search.

search

Search indexed public-source records by keyword, returning facts, synthesized extracts, and citations.

diff_item

Compare document versions and inspect additions, deletions, and change summaries.

get_item

Retrieve full items with titles, dates, source URLs, and segment windows for close reading.

get_item_neighbors

Walk previous and next comparable items around a source document, nearest first.

get_entity

Fetch canonical entity detail, including aliases, roles, and cross-source coverage counts.

get_entity_items

Traverse items linked to an entity, with optional in-scope query filtering and source-aware access handling.

search_entities

Resolve canonical entities by name or alias and see which people or institutions the corpus already links.

Reference

Tool Reference

Tool	Purpose	Required input	Returns
`get_stats`	Inspect corpus coverage, counts by type, and language/date range metadata.	none or `source`	coverage and type counts
`list_items`	Browse corpus items by source, subtype, language, and date without running text search.	`source`	item list with metadata
`search`	Search indexed public-source records by keyword, returning facts, synthesized extracts, and citations.	`query`	matched documents and segments
`diff_item`	Compare document versions and inspect additions, deletions, and change summaries.	`source_id`	document diff
`get_item`	Retrieve full items with titles, dates, source URLs, and segment windows for close reading.	`source_id`	full item with segments
`get_item_neighbors`	Walk previous and next comparable items around a source document, nearest first.	`source_id`	adjacent item lists
`get_entity`	Fetch canonical entity detail, including aliases, roles, and cross-source coverage counts.	`entity_id`	entity detail
`get_entity_items`	Traverse items linked to an entity, with optional in-scope query filtering and source-aware access handling.	`entity_id`	items linked to entity
`search_entities`	Resolve canonical entities by name or alias and see which people or institutions the corpus already links.	`query`	entity matches

This product is in beta and provided for informational purposes only. It does not constitute investment, trading, legal, or tax advice. Content is generated from public-source materials and may contain errors, omissions, or outdated information. Data and analysis are provided without warranty of accuracy or completeness, and no liability is accepted for losses arising from use of this information. Users are responsible for their own evaluation, decisions, and risk management.