What the corpus is
research_corpusSize, span (1736-1963), decade histogram, quality and caution shares, and coverage of Chronicling America (English-language titles are the target). Call once before searching so a negative result means something. Free.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "research_corpus"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| docs | Yes | Event docs in the index | |
| issues | Yes | Issues with at least one component in the corpus | |
| states | Yes | ||
| titles | Yes | Newspaper titles (LoC lccns) with processed issues | |
| decades | Yes | docs per decade, e.g. {"1850": 24387} | |
| year_max | Yes | ||
| year_min | Yes | ||
| components | Yes | Components (a multi-event notice is one component, N docs) | |
| last_ingest | Yes | ||
| caution_docs | Yes | docs flagged offensive by modern standards | |
| quality_tiers | Yes | docs per tier (green / yellow / red). red is reported here and withheld everywhere else: never a hit, a facet bucket or a fetchable component. Every other count is the served corpus. | |
| embed_worthy_docs | Yes | docs with a vector (semantic search can see them) | |
| denominators_computed_at | No | Counted live, cached ~1 h | |
| issues_in_chronicling_america | No | Every issue LoC has published for an ENGLISH-LANGUAGE title — the target corpus and the coverage denominator; refreshed with the crawler | |
| coverage_of_chronicling_america | No | issues / issues_in_chronicling_america | |
| issues_in_chronicling_america_all_languages | No | Same across all languages, for reference (English is > 95 %) |