API reference Findings and fixes
Findings and fixes
Checks, newest first
GET /checks
Checks, newest first; open=true keeps the ones not done
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | |
fix_group | query | string | |
open | query | string | |
limit | query | string | |
offset | query | string |
Responses
How each fix group or asset can be checked again
GET /checks/options
How each fix group (fix_group) or asset (asset_id) can be checked again: on_demand when a check gets a fresh report within minutes, else reason (removed, agent, upload, not_found), with method, host_name and the latest check (last_check). Repeat a parameter or separate values with commas; at most 50.
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
fix_group | query | string | |
asset_id | query | string |
Responses
One check
GET /checks/{id}
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
List config findings
GET /config-findings
Risky settings collectors found in files: AI coding-agent hooks and MCP start commands that download and run code, SSH keys and settings that give root an unrestricted shell, GitHub Actions that run third-party actions by a movable tag. Each has why, fix and, where there is one, doc_url. No secret is stored
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | open (the default), resolved or all |
severity | query | string | high, medium, low or info, comma-separated |
category | query | string | agent_hook, ssh or ci, comma-separated |
asset_id | query | string | Limit to one asset |
limit | query | string | Page size |
offset | query | string | Page offset |
Responses
Counts of open config findings
GET /config-findings/summary
Open config findings by severity and by category, and how many assets were checked at all
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | Limit to one asset |
Responses
Dismiss a config finding
POST /config-findings/{id}/dismiss
Dismiss a config finding (a key or a hook that is meant to be there), with an optional note. It stays dismissed whatever later scans see
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Open a dismissed config finding again
POST /config-findings/{id}/undismiss
Open a dismissed config finding again. 409 for one that is not dismissed
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Dashboard SLA split and 30-day trend
GET /dashboard/trend
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
tz | query | string | IANA time zone where today ends (default UTC) |
Responses
Queue a full re-evaluation
POST /evaluate
Request body
application/json, required: object
Responses
Vulnerability and end-of-life findings, SOS, priority, known-exploited and most severe first
GET /findings
Vulnerability and end-of-life findings, SOS, priority, known-exploited and most severe first. new_within=N keeps findings first seen in the last N days, resolved_within=N those resolved in the last N days (with status=resolved), resolution_kind (comma-separated: remediated, removed, superseded, advisory_withdrawn, advisory_changed, not_evaluated, other) those closed for that reason, and sort=asset orders by asset name first. epss_min (0 to 1) keeps findings one of whose CVEs has at least that EPSS. Each finding carries epss, epss_percentile, epss_date, epss_model_version and epss_cve (null without a score): the highest EPSS among the finding's CVEs (its key, and the CVEs its advisories list as aliases or upstream ids), and the CVE it is for. Also filters by priority, state, sos, overdue, assignee (assignee_user, assignee_team, assignee=none), fix_group, incident_id and second_check; paged with limit and offset
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | |
kind | query | string | |
severity | query | string | |
verification | query | string | |
asset_id | query | string | |
component_id | query | string | |
environment | query | string | |
known_exploited | query | string | |
epss_min | query | string | Vulnerability findings whose CVE has a FIRST EPSS of at least this probability, 0 to 1 (0.1 is 10%) |
q | query | string | |
key | query | string | |
new_within | query | string | |
resolved_within | query | string | |
resolution_kind | query | string | |
sort | query | string | |
priority | query | string | P1 to P4, comma-separated |
state | query | string | active, new, acknowledged, in_progress, suppressed, risk_accepted or regressed |
sos | query | string | true: SOS findings only |
overdue | query | string | true: past their fix deadline |
due_before | query | string | Active findings whose fix-by date is before this time (RFC 3339) |
due_after | query | string | Active findings due at or after this time (RFC 3339) |
assignee_user | query | string | Assigned to this user id |
assignee_team | query | string | Assigned to this team id |
assignee | query | string | none: unassigned findings |
fix_group | query | string | In this fix group (its id) |
incident_id | query | string | In this incident |
second_check | query | string | agreed, disagreed or could_not_check |
limit | query | string | Page size (default 100, at most 500) |
offset | query | string | Items to skip |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FindingList |
4XX | An error | Error |
5XX | An error | Error |
Apply a workflow action to up to 500 findings
POST /findings/actions
The same action on up to 500 findings (ids)
Request body
application/json, required: object
Responses
Findings grouped by asset, paged by asset: each asset's count and its most urgent findings
GET /findings/by-asset
The findings that match the GET /findings filters, grouped by asset and paged by asset in order of asset name. Each item is an asset: asset_id, asset_name, asset_kind, total (its findings that match) and findings, the first per_asset (1 to 50, default 5) in the GET /findings order. total at the top counts assets, findings_total findings; limit and offset count assets. Narrowed to the caller's projects
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | |
kind | query | string | |
severity | query | string | |
verification | query | string | |
asset_id | query | string | |
component_id | query | string | |
environment | query | string | |
known_exploited | query | string | |
epss_min | query | string | Vulnerability findings whose CVE has a FIRST EPSS of at least this probability, 0 to 1 (0.1 is 10%) |
q | query | string | |
key | query | string | |
new_within | query | string | |
resolved_within | query | string | |
resolution_kind | query | string | |
priority | query | string | P1 to P4, comma-separated |
state | query | string | active, new, acknowledged, in_progress, suppressed, risk_accepted or regressed |
sos | query | string | true: SOS findings only |
overdue | query | string | true: past their fix deadline |
due_before | query | string | Active findings whose fix-by date is before this time (RFC 3339) |
due_after | query | string | Active findings due at or after this time (RFC 3339) |
assignee_user | query | string | Assigned to this user id |
assignee_team | query | string | Assigned to this team id |
assignee | query | string | none: unassigned findings |
fix_group | query | string | In this fix group (its id) |
incident_id | query | string | In this incident |
second_check | query | string | agreed, disagreed or could_not_check |
limit | query | string | Page size (default 100, at most 500) |
offset | query | string | Items to skip |
per_asset | query | integer |
Responses
Counts by severity, kind, verification, priority and state, plus SOS, overdue, unassigned, stale feeds,…
GET /findings/summary
Counts by severity, kind, verification, priority and state, plus SOS, overdue, unassigned, stale feeds, evaluation gaps and resolved_7d (resolved in the last 7 days)
Responses
What was resolved, and what is still open, since a time
GET /findings/verify
What changed since since (RFC 3339 time or YYYY-MM-DD): findings resolved since then with their cause, resolution_kind and whether it counts_as_remediation (the totals remediated and closed_not_remediation, and resolved_by_kind), findings that existed then and are still_open, opened_since, and each asset's last complete collection. asset_rescanned_since tells "fixed" from "not scanned again yet". Lists are capped at limit (default 100, max 500) with a total
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
since | query | string | |
asset_id | query | string | |
limit | query | string |
Responses
One finding, with every source advisory in its group and, while it is open, its upgrade recommendation
GET /findings/{id}
One finding (with resolution_kind and resolution_detail once resolved), with every source advisory in its group and, while it is open, its upgrade recommendation
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FindingDetail |
4XX | An error | Error |
5XX | An error | Error |
Apply a workflow action to a finding
POST /findings/{id}/actions
Workflow action, see below
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json, required: object
Responses
The proof behind a closed finding (verified closure): verification (verified_fixed, verified_removed, disappeared_unverified or closed_not_remediation), the version before and after, the check that saw it, who fixed it, and time to fix against the fix-by date. 404 while the finding is open
GET /findings/{id}/closure
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
A finding's workflow history
GET /findings/{id}/events
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
Why the finding matched (why, as in GET /findings/{id}) and why it has its priority
GET /findings/{id}/explain
Why the finding matched (why, as in GET /findings/{id}) and why it has its priority: risk (priority, score, each factor with its points, policy_version: the risk policy that produced this priority, current_policy_version and policy_changed, the current thresholds, the SLA for the priority, deadlines, and epss: FIRST's EPSS score, percentile, model and date for the CVE, the policy threshold and whether and why it counts, also under why.epss), the stored evidence, the matched advisories (id, feed, severity, dates, an osv.dev link), the freshness of the feeds they came from, and the asset's last (complete) collection time
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
The upgrade recommendation for the installed copy the finding is on, covering every open finding on it
GET /findings/{id}/upgrade
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
Fixes (fix groups), most urgent first
GET /fix-groups
Fixes (fix groups), most urgent first. bucket is focus (a fix exists, SOS or P1, on a production asset or one with no environment), no_fix, old_kernels, non_production or lower_priority; every group is in exactly one, checked in that order. priority, overdue, exploited and assignee=me narrow the list. group_by is fix (the same change in several places is one entry with places: group, asset, app directory and finding count per place; counts are added up, unique_issues is not) or asset (one entry per asset); the default is the workspace's setting (/settings/fix-grouping)
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | |
kind | query | string | |
environment | query | string | |
q | query | string | |
bucket | query | string | |
priority | query | string | Most urgent issue: P1, P2, P3 or P4, several separated by commas |
overdue | query | string | true: fixes with an active issue past its fix deadline |
exploited | query | string | true: fixes with a known exploited issue |
assignee | query | string | me: fixes with an open issue assigned to the caller One of me. |
group_by | query | string | One of fix, asset. |
limit | query | string | Page size (default 100, at most 500) |
offset | query | string | Items to skip |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixGroupList |
4XX | An error | Error |
5XX | An error | Error |
Fix groups and findings per kind and bucket
GET /fix-groups/summary
Groups and findings per kind, plus focus: the number of focus groups and their findings, groups and findings per other bucket (hidden), and the assets with focus groups. by_priority counts fixes by their most urgent issue, overdue those with an active issue past its fix deadline, exploited those with a known exploited issue. Counts follow group_by as GET /fix-groups does: grouped by fix, the same fix in several places counts once
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | |
group_by | query | string | One of fix, asset. |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixPlanSummary |
4XX | An error | Error |
5XX | An error | Error |
Where the work on each listed fix stands
GET /fix-groups/work
ids: up to 500 fix group ids, separated by commas. One item per group with open issues the caller can see: group_id, active, due (earliest fix deadline of an active issue), overdue, in_progress, regressed, owners (assignee emails or team names, most issues first, at most five), unassigned and mine (assigned to the caller)
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
ids | query | string | Fix group ids, separated by commas |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixWorkList |
4XX | An error | Error |
5XX | An error | Error |
One fix group with its findings, paged
GET /fix-groups/{id}
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string | |
limit | query | string | |
offset | query | string |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixGroupDetail |
4XX | An error | Error |
5XX | An error | Error |
Mark a fix as applied and request a check
POST /fix-groups/{id}/applied
"I applied this": the fix's active issues go to in_progress with the reason "applied, waiting for a check", and a check is requested. Nothing is resolved by this call. Returns updated and check
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixApplied |
4XX | An error | Error |
5XX | An error | Error |
Check a fix again
POST /fix-groups/{id}/check
Check again the fix: asks the asset's agent (or a server-side scan) for a fresh report. 201 with the new check, or 200 with the check already waiting for a report
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
The approval inbox
GET /ignore-requests
The approval inbox. status is pending (default), approved, declined or all. Security and admin users see every request; others see their own. can_decide says which
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | |
limit | query | string | |
offset | query | string |
Responses
Ask to ignore issues
POST /ignore-requests
Ask to ignore issues: finding_ids (up to 2,000 open findings) or fix_group (its active issues), with reason, scope, expiry, until, note and mitigation as for the ignore action. Validated like an ignore; 201 with the request
Request body
application/json, required: object
Responses
One request (its requester, or security and admin users)
GET /ignore-requests/{id}
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
Approve or decline a request to ignore issues
POST /ignore-requests/{id}/decision
{"decision": "approve" | "decline", "note": "..."}. Approval ignores the request's issues that are still open and active, as the decider; 409 when already decided
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json, required: object
Responses
Ignore rules in force (or every one with status=all), with the open findings each ignores
GET /ignore-rules
Ignore rules in force (or every one with status=all), with the open findings each ignores
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | |
limit | query | string | Page size (default 100, at most 500) |
offset | query | string | Items to skip |
Responses
End an ignore rule
DELETE /ignore-rules/{id}
End a rule: the open findings it ignores return to new
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
Incidents: findings grouped by issue and owner team
GET /incidents
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
status | query | string | |
kind | query | string | |
priority | query | string | |
sos | query | string | |
team_id | query | string | |
escalated | query | string | true: escalated incidents only |
limit | query | string | Page size (default 100, at most 500) |
offset | query | string | Items to skip |
Responses
An incident with its findings (and deliveries, for security and above)
GET /incidents/{id}
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Responses
Apply a workflow action to every open finding of an incident
POST /incidents/{id}/actions
Apply an action to every open finding in the incident
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json, required: object
Responses
Leaked keys found in files, most severe first
GET /leaks
Keys, tokens and passwords collectors found written in assets' files: kind of key, file and line, setting name, fingerprint, masked preview, severity and exposure. The value is never stored. Each has its state, owner, due date and rotation
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | Limit to one asset |
status | query | string | open (default), resolved or all |
severity | query | string | critical, high, medium or low |
state | query | string | needs_rotation, waiting_check, still_found, verified_gone, dismissed or overdue |
assignee_user | query | string | Assigned to this user id |
assignee_team | query | string | Assigned to this team id |
assignee | query | string | none: with no owner |
limit | query | string | Page size |
offset | query | string | Page start |
Responses
Counts of leaked keys by what needs doing
GET /leaks/summary
How many leaked keys need rotation, are still found after rotation, are overdue, wait for a check, were verified gone in the last 30 days and have no owner
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
asset_id | query | string | Limit to one asset |
severity | query | string | critical, high, medium or low, comma-separated |
Responses
Assign a leaked key
POST /leaks/{id}/assign
Set who rotates a leaked key: user_id and/or team_id; 0 clears one, one left out stays
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Check a leaked key again
POST /leaks/{id}/check-again
Ask for the leak's asset to be scanned again, as Check again does for an asset
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Dismiss a leaked key
POST /leaks/{id}/dismiss
Dismiss a leaked key (a test key kept on purpose, or one already rotated), with an optional note. It stays dismissed whatever later scans see
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Mark a leaked key as rotated
POST /leaks/{id}/rotated
Record that the key was revoked and replaced, with an optional note; with check true, also ask for a check. It stays open until a complete scan no longer sees the key
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
Open a dismissed leaked key again
POST /leaks/{id}/undismiss
Open a dismissed leaked key again. One resolved as gone stays resolved until a scan sees it
Parameters
| Parameter | In | Type | Description |
|---|---|---|---|
id required | path | string |
Request body
application/json: object
Responses
The risk policy: priority weights, thresholds, deadlines and SOS rules
GET /risk-policy
Priority weights, thresholds, deadlines and SOS rules, with the policy version. due_dates is patch_calendar when non-urgent findings are due at their asset's next patch window, set by patch_calendar (test_days_after_patch_tuesday 0-20, default 1; prod_days_after_test 0-28, default 7; timezone, default UTC), and is left out for sla, the default (risk.md). unmaintained_months (6-120, default 24, left out while it is the default) is how long without a release makes a package unmaintained in a no_fix group's no_fix_advice; a change applies at the next action plan rebuild, within about a minute. PUT replaces the policy (fields left out take their defaults; unknown fields are rejected), returns version, previous_version and the changed fields, and writes a risk_policy.update audit event holding the policy before and after. GET includes last_change (who, when, versions, changed fields; null when never changed)
Responses
Replace the risk policy
PUT /risk-policy
Priority weights, thresholds, deadlines and SOS rules, with the policy version. due_dates is patch_calendar when non-urgent findings are due at their asset's next patch window, set by patch_calendar (test_days_after_patch_tuesday 0-20, default 1; prod_days_after_test 0-28, default 7; timezone, default UTC), and is left out for sla, the default (risk.md). unmaintained_months (6-120, default 24, left out while it is the default) is how long without a release makes a package unmaintained in a no_fix group's no_fix_advice; a change applies at the next action plan rebuild, within about a minute. PUT replaces the policy (fields left out take their defaults; unknown fields are rejected), returns version, previous_version and the changed fields, and writes a risk_policy.update audit event holding the policy before and after. GET includes last_change (who, when, versions, changed fields; null when never changed)
Request body
application/json, required: object
Responses
Preview a risk policy's effect without saving it
POST /risk-policy/preview
Validates a proposed policy like PUT and returns its effect on the workspace's open findings without saving anything: before / after counts per priority, sos_before / sos_after, priority_changed (raised, lowered), sos_changed, score_changed, stored_differs, and sample (at most sample_limit, 50) of changed findings with factor deltas (risk.md)
Request body
application/json, required: object
Responses
The independent second check of high-impact findings
GET /second-checks
The independent second check of high-impact findings: open findings agreed, disagreed and could_not_check, pending (covered, not checked yet), logged_30d and the 50 latest logged disagreements with their differences and inputs. GET /findings?second_check=disagreed (or agreed, could_not_check) lists the findings; every finding carries second_check and second_check_reason, and GET /findings/summary counts second_check_disagreed. A disagreement never changes a finding's verification, priority, SOS or alerts. Narrowed to the caller's projects
Responses
How fix lists group the same fix in several places
GET /settings/fix-grouping
How fix lists show the same fix in several places: fix_grouping is fix (one entry per fix listing its places; the default) or asset (one entry per asset). PUT writes a settings.update audit event
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixGrouping |
4XX | An error | Error |
5XX | An error | Error |
Change how fix lists group the same fix
PUT /settings/fix-grouping
How fix lists show the same fix in several places: fix_grouping is fix (one entry per fix listing its places; the default) or asset (one entry per asset). PUT writes a settings.update audit event
Request body
application/json, required: FixGrouping
| Field | Type | Description |
|---|---|---|
fix_grouping required | string | fix: one entry per fix, listing its places (the default); asset: one entry per asset and fix. One of fix, asset. |
Responses
| Status | Meaning | Body |
|---|---|---|
2XX | Success | FixGrouping |
4XX | An error | Error |
5XX | An error | Error |
The organisation's two-factor policy, and who would need to set up a factor
GET /settings/security
Responses
Change the organisation's two-factor policy
PUT /settings/security
Request body
application/json, required: object