AI
Connect ChatGPT or Claude to Better EDGAR over the Model Context Protocol.
Pick your client
Each tab has the exact steps and copy-paste values for that client.
Add to ChatGPT
ChatGPT Business, Enterprise, or Edu (web only). First enable Developer Mode: Settings → Apps → Advanced Settings → toggle Developer mode. Then go to Settings → Apps → Create, paste the URL below, set Authentication to OAuth, and click Create. No key needed.
On Enterprise/Edu, an admin must first grant access via Permissions & Roles → Connected Data.
https://better-edgar.com/mcp
MCP server reference
Same filings, packaged as tool calls. Streamable HTTP (JSON-RPC 2.0).
- Endpoint
https://better-edgar.com/mcp- Transport
- Streamable HTTP (POST JSON-RPC 2.0). Protocol version
2025-11-25. - Auth
- OAuth 2.1 with Dynamic Client Registration.
- Capabilities
tools/list,tools/call,prompts/list,prompts/get.
Tools
| Name | Summary |
|---|---|
| search_filings | Search EDGAR filings by issuer, forms, topics, date, and saved watchlist. |
| search_news | Search AI-generated news items across filings. |
| search_issuers | Resolve a fuzzy ticker or name to a CIK. |
| get_insider_transactions | Search Form 4 insider transactions by issuer, owner, date, and trade type. |
| meta | Enum reference — forms, sectors, industries, exchanges (with q filter). |
| list_watchlists | List your watchlists. |
| get_watchlist | Fetch a watchlist and its member companies. |
| manage_watchlist | Create, edit, delete, or change companies in your watchlists. |
| get_activity | Paginated issuer-grouped dashboard activity with filters and cursors. |
| dashboard_overview | Agent-facing sector pulse with activity density, price movers, and drilldowns. |
| dashboard_sector | Drill into one sector with industries, movers, and issuer activity. |
| dashboard_industry | Drill into one industry with issuer activity and price movers. |
| dashboard_issuer | Dashboard-style recent activity for one ticker or CIK. |
| get_latest_news | Latest generated EDGAR news by ticker, sector, industry, topics, or keyword. |
| list_feeds | List your feeds. |
| get_feed_matches | Fetch the latest matches for a named feed. |
| get_filing | Retrieve a filing or one of its exhibits. |
| filing_profile | Profile a filing: ToC, table counts, notable tables, exhibits, and XBRL coverage. |
| search_filing | Search ranked snippets across text and tables inside a specific filing. |
| read_filing | Read a targeted section, chunk, or table from a filing. |
| search_filing_tables | Find filing tables by keyword, table text, title, key terms, or concepts. |
| get_filing_table | Return a parsed filing table as semantic rows, markdown, facts, or all views. |
| xbrl_search_concepts | Search XBRL concepts with table and label context. |
| xbrl_get_facts | Fetch targeted inline XBRL facts by concept, query, table, or fact kind. |
| xbrl_compare_periods | Compare one XBRL concept across periods and dimensions. |
| help | Concept reference — agent workflow, CIK, forms, pagination, market cap, XBRL. |
Full tool schemas and examples are served via tools/list on the live endpoint. Every tool
is read-only except manage_watchlist, which
can change only the signed-in user's watchlists and is explicitly marked destructive.
Prompts
| Name | Summary |
|---|---|
| start_here | Lite request-triage guide: classify the ask, resolve scope, clarify only high-impact gaps, then plan tool use. |
| getting_started | Guided walkthrough of common BetterEDGAR MCP workflows. |
| analyst_retrieval_testset | Representative analyst questions for testing filing retrieval workflows. |