listFindings
List and filter Mallory findings, returning lean summaries as {items,total,offset,limit}. Each item includes: title, status, severity, confidence, the owning definition's slug/title/category, asset type and identifier, resolution_reason, the thread_uuid of the investigation that produced it, and timestamps; the detail object adds body / summary / remediation / evidence, any linked external tickets, and entities — the threat actors, vulnerabilities and stories this finding resolved to, which is the pivot back into Mallory's intel graph. Returns every status by default — pass status=open (or status__in=open,fixed) to narrow. Sort by created_at, updated_at, status, title, severity, or asset_identifier; severity sorts INFO < LOW < MEDIUM < HIGH < CRITICAL, so desc puts Critical first. Notable fieldFilters: status, severity, definition_slug, definition_category, asset_type, asset_identifier, title, resolution_reason (also resolution_reason__isnull=true for closed findings with no reason recorded), thread_uuid to scope to one investigation. Enum values are case-sensitive and each field has its own set: severity is UPPERCASE (CRITICAL, HIGH, MEDIUM, LOW, INFO); status is lowercase and has exactly three values (open, fixed, dismissed); resolution_reason is lowercase and separate from status (false_positive, accepted_risk, not_applicable, duplicate, fixed_externally, other); definition_category is lowercase-hyphenated (e.g. public-exposure). Send values, not enum member names. definition_category is the exception: equality and membership ONLY (bare, __in, __not_in) — the __contains/__ilike/__neq the other fields accept are dropped on it, and you get the unfiltered list back. asset_type is a free-text column in provider.service.object form (e.g. aws.s3.bucket, not s3), so a partial value matches nothing unless you use asset_type__ilike. Field operators append to a field name with __: a bare field name is exact match, and the string/enum fields (status, severity, title, definition_slug, asset_type, asset_identifier, resolution_reason) also take __neq, __in, __not_in, __contains, __not_contains, __ilike. This endpoint has NO __like. Timestamps are created_at / updated_at with __gte and __lt ONLY, a half-open range — __gt, __lte and a bare created_at are not parameters at all. An operator this endpoint does not define is dropped silently rather than rejected, so guessing one returns a WIDER result set that looks filtered. Pass an item's identifier to getFinding for the full detail object.
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": "listFindings"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| sort | No | Field to sort by (see this tool's sort fields). | |
| limit | No | Max items to return (1-100, default 25). | |
| order | No | Sort order (default: desc). | |
| filter | No | Full-text search over the finding's title, asset identifier, body, summary, remediation and asset details. Supports web-search query syntax, and AND-combines with the other filters. | |
| offset | No | Items to skip, for pagination. | |
| fieldFilters | No | Advanced structured filters as a map of query-param name -> value. Use a bare field for exact match, or a field__operator key for ranges/sets. Field operators append to a field name with `__`: a bare field name is exact match, and the string/enum fields (status, severity, title, definition_slug, asset_type, asset_identifier, resolution_reason) also take __neq, __in, __not_in, __contains, __not_contains, __ilike. This endpoint has NO __like. Timestamps are created_at / updated_at with __gte and __lt ONLY, a half-open range — __gt, __lte and a bare created_at are not parameters at all. An operator this endpoint does not define is dropped silently rather than rejected, so guessing one returns a WIDER result set that looks filtered. |