AUMSearch for developers
Query adviser, private-fund, Form D, and Form 13F records through the API or receive scheduled files in your environment. Both preserve identifiers, dates, and source context.
/v1/managers?query=ICONIQ&limit=2{
"data": [{
"id": "11111111-1111-4111-8111-111111111111",
"display_name": "ICONIQ Capital, LLC",
"provenance": {
"source": "FORM_ADV",
"source_url": "https://adviserinfo.sec.gov/..."
}
}],
"meta": { "request_id": "req_...", "has_more": true, "next_cursor": "..." },
"links": { "self": "...", "next": "..." }
}Choose by workload
The interfaces share the same source-first model, but they solve different operational problems. Start with the way your system needs to consume and refresh the data.
REST API
The published specification is the source of truth for paths, parameters, schemas, headers, and errors. The overview below shows the behavior an integration should be designed around.
Authenticated REST requests use an issued API key with the api:read scope. Keys are sent in the Authorization header and can be rotated without changing an integration.
List resources return data, meta, and links. Pass meta.next_cursor back as cursor instead of constructing offsets or depending on row position.
Successful responses expose ETag and X-AUMSearch-Data-Version headers so a client can identify unchanged results and trace a response to its source-data revision.
Account-specific limits are communicated through RateLimit and X-RateLimit headers. A 429 response includes Retry-After for bounded backoff.
Errors return an HTTP status, an explicit error code, a human-readable message, and request context instead of an unstructured HTML response.
Normalized records retain the relevant source, source URL, filing identity, report period, and source timestamps when those values apply.
https://aumsearch.com/v1Bearer <api-key>Search firms, retrieve a sourced profile, follow filing history, and compare reported periods.
/v1/managersSearch and filter investment managers/v1/managers/{id}Retrieve a manager profile and identifiers/v1/managers/{id}/filingsList a manager’s sourced filing history/v1/managers/{id}/comparisonsCompare adjacent reported Form 13F periodsSearch filing metadata or retrieve the structure, source context, and parser warnings for one filing.
/v1/filingsSearch filing metadata across sources/v1/filings/{id}Retrieve a filing, report structure, and warningsQuery reported Form 13F positions by manager, security, report period, identifier, or change type.
/v1/holdingsQuery reported institutional holdingsSearch normalized securities, resolve source-native identifiers, and inspect identifier history.
/v1/securitiesSearch normalized securities by canonical name or identifier/v1/securities/resolveResolve a dated source-native security identifier/v1/securities/{id}Retrieve a normalized security recordRead public service health and dataset statistics, or use the credentialed source-level coverage model.
/v1/statusInspect service health and all dataset statistics/v1/datasets/statsList record counts and freshness for every dataset/v1/datasets/{id}/statsRetrieve record count and freshness for one dataset/v1/sources/statusInspect source freshness and coverageFailure behavior
Use the HTTP status for transport behavior and the machine-readable code for application logic. Record the request identifier when escalating a failed request.
HTTP/1.1 429 Too Many Requests
Retry-After: 24
RateLimit-Remaining: 0
{
"data": null,
"error": {
"code": "RATE_LIMITED",
"message": "The API key rate limit was exceeded."
},
"meta": { "request_id": "req_..." },
"links": {}
}Data feeds
A feed is defined by more than a file format. The delivery profile records what is included, how change is represented, where it is sent, and how both systems confirm a successful load.
s3://research-data/aumsearch/advisers/Dataset families, fields, historical window, and any source-specific coverage limits.
A full replacement snapshot, bounded changes since the prior successful delivery, or both.
A delivery schedule aligned to source publication patterns and the workflow that consumes the files.
SFTP, an Amazon S3 bucket, or an Azure Blob Storage container controlled by the agreed delivery profile.
Format, encoding, compression, field types, null behavior, naming, and schema-version policy.
Source cutoff, delivery window, manifest fields, validation, retries, retention, and support path.
Delivery manifest
A delivery manifest should let a pipeline identify the feed, dataset, schema, source cutoff, files, and record totals before promoting data downstream. Exact fields and file conventions are recorded in the delivery profile.
{
"delivery_id": "del_...",
"dataset": "adviser_firms",
"schema_version": "...",
"source_cutoff": "...",
"files": [{ "name": "...", "records": "..." }]
}Dataset contracts
Each guide documents representative normalized columns, native identifiers, filing and event dates, amendment behavior, joins, and interpretation limits. Feed schemas preserve as-filed values and provenance beside normalized fields.
Connect firms and representatives to their registrations, narrative disclosures, private funds, service providers, and exempt offerings.
Adviser firms · Form ADV filings · ADV-W · Ownership · DisclosuresPrivate fund IDs · Fund types · Gross assets · Service providers · Form D linksIndividual CRD · Firm history · State registrations · Qualifications · DisclosuresPart 2A brochures · Part 2B supplements · Form CRS · Document versions · Extracted sectionsForm D/D-A filings · Issuers · Offerings · Recipients · AmendmentsMove from a registrant and series to its registration lifecycle, portfolio holdings, share classes, and source filings.
Form N-1A · Form 485 · Series IDs · Class IDs · ProspectusesFund portfolios · Position values · Debt terms · Derivatives · CounterpartiesStudy periodic holdings and the identifiers connecting managers, issuers, securities, filings, and fund portfolios.
Form 13F filings · Reported holdings · Managers · Report periods · AmendmentsCUSIP · Ticker · FIGI · Issuer · Security classImplementation boundary
This lets engineering begin against a real data model without turning an account-specific quota, cadence, or warehouse decision into a universal product promise.
Share the resources or datasets, historical depth, expected volume, users, and destination.
Confirm credentials, filters, pagination or file loading, provenance, and failure handling against a real sample.
Document owners, quotas or delivery windows, monitoring, retries, schema changes, and the support path.
Build with AUMSearch
Tell us what you are building, which records it needs, and how the data should move through your system.