OpenFilings
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DART_API_KEY | No | API key for DART (South Korean filings) | |
| EDINET_API_KEY | No | API key for EDINET (Japanese filings) | |
| OPENFILINGS_OCR_DPI | No | OCR DPI setting | 200 |
| OPENFILINGS_DATA_DIR | No | Directory for SQLite cache (defaults to .openfilings) | |
| OPENFILINGS_OCR_MODE | No | OCR mode: auto, never, or always | auto |
| OPENFILINGS_CACHE_MAX_MB | No | Maximum cache size in MB | 512 |
| OPENFILINGS_OCR_LANGUAGE | No | OCR language(s) (e.g., eng, eng+fra) | eng |
| OPENFILINGS_OCR_MAX_PAGES | No | Maximum pages for OCR | 250 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| companies_searchB | Find companies using compact metadata; start most company workflows here. |
| filings_listB | List compact filing metadata without downloading document content. |
| disclosures_searchA | Full-text search across every issuer's disclosures, not scoped to one company. Only fca_nsm (headline search) and cvm (its yearly filing index) support this. |
| company_factsA | Merge a company's most recent structured filings into one multi- period fact series per line item, instead of one filing at a time. |
| data_quality_reportB | Report extraction provenance, confidence, and validation for one filing. |
| financials_queryB | Return only selected multi-period financial facts with provenance. |
| historical_backfillB | Download available UK/ESEF structured filing history into local storage. |
| historical_facts_queryC | Query locally stored filing facts with restatement-aware semantics. |
| company_research_briefA | Return recent filings and a compact three-period financial profile. Ownership and director-dealing data stay in their dedicated tools because coverage differs by source; missing data must not imply no activity. |
| companies_compareB | Compare one normalized fact across two to twenty UK/ESEF issuers. |
| filings_diffB | Compare normalized facts in two filings and return changed values. |
| watchlist_checkC | Statelessly return filings published on or after |
| major_holders_listC | List structured major-holder records for a UK or Brazilian issuer. |
| major_holders_searchC | Search holder positions across UK notifications and Brazilian FRE data. |
| insider_dealings_listB | List UK MAR director/PDMR/PCA dealings parsed into structured fields. |
| sedar_filing_importC | Import one official SEDAR+ generated PDF URL into the local cache. |
| filing_outlineB | Return headings and sizes only, allowing targeted section selection. |
| filing_sectionsB | Compatibility outline tool; returns headings without section bodies. |
| filing_readB | Read one named filing section through a bounded, paginated response. |
| filing_searchB | Search one filing and return ranked, bounded excerpts. |
| filing_markdownA | Read a bounded Markdown page; prefer outline, search, or section tools. |
| filing_financialsC | Return selected statements with bounded periods and line items. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 22 tools
Most tools target distinct resources or actions, but filing_outline and filing_sections are nearly identical, both returning headings without section bodies. Additionally, filing_read, filing_markdown, and filing_search could be confused if descriptions are not carefully read, though their purposes are subtly different.
Tool names mostly follow an object-first pattern (e.g., filings_list, companies_search), but the action placement is inconsistent—sometimes at the end (filings_diff, financials_query) and sometimes absent or noun-based (filing_outline, company_research_brief, company_facts). This mix makes the naming less predictable than a uniform verb_noun scheme.
22 tools is on the higher end but appropriate for a comprehensive filings domain that spans company research, filing retrieval, structured financial queries, ownership data, and historical backfill. The count is not excessive given the breadth, though one redundant tool pair slightly inflates it.
The tool set covers the core workflows: locating companies, listing and reading filings, extracting structured financial facts, comparing filings, and monitoring new submissions. It also includes an import tool for external filings. Minor gaps exist (e.g., no full-text search for insider dealings), but the surface is largely complete for the intended use.