> ## Documentation Index
> Fetch the complete documentation index at: https://doc.astreus.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Search and entity resolution

> Resolve canonical identifiers and find attributable passages across the held corpus.

## Resolve an entity

```text theme={null}
GET /api/v1/search?q=DEWA&scope=all&limit=5
```

The default curated scope supports the human terminal's company, series and
country navigation. `scope=all` adds commodities, FX pairs, people, conflicts
and organisations and includes national/regional macro series with stable IDs.
`country=AE` narrows macro series to a country and its regions. `limit` is per
group, bounded to 1–20.

Use the returned company `cik`, including a letter-prefixed platform key, for
subsequent requests. A ticker alone can collide across markets. An empty bucket
does not prove that an entity never existed; covered/published inventory bounds
the response, and a search bucket may fail independently.

The CLI's formatted `search` command remains curated. For full programmatic scope:

```bash theme={null}
astreus --json api list --search search
# Use the returned operation ID:
astreus --json api call get-search --query q=DEWA --query scope=all
```

MCP `search` requests the programmatic scope and returns exact identifiers.

## Search source passages

`GET /api/v1/transcripts/search` supports websearch expressions: quoted phrases,
`OR`, and `-exclusion`. Filter by company or basket, dates, speaker/role, section,
topic or explicit lexical theme. `sort=relevance` ranks matches; the default is
newest first. `includeContext=1` requests surrounding speaker turns or exchanges.
Use the returned cursor to continue the same query.

Passages retain issuer, speaker, call date and character spans. Retrieve context
before quoting. Theme expansion is lexical matching, not an inferred semantic
claim. `/summary`, `/facets` and `/topics` provide counts, filter options and
inspectable topic comparisons; these have coverage and result-cap caveats.

Use `search_transcripts`, `analyze_transcript_topics` or `get_mentioned_by` over
MCP, or `astreus transcript-search` and `astreus mentioned-by` in the CLI.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.