Dome Systems

dome audit

Query, stream, get, export, catalog, and SIEM-stream audit events

dome audit reads the evidence Dome records for every governed action. Query or tail events, follow one activity chain end to end, browse the event catalog, and export to a file or a SIEM destination. Refer to Audit for what each event carries.

Command
Query eventsdome audit query
Stream eventsdome audit stream
Get eventdome audit get [event-id]
Export eventsdome audit export
Catalogdome audit catalog list / show
SIEMdome audit siem enable / disable / resume / status / list
List chainsdome audit chains
Show chaindome audit chain [activity-id]

All commands scope the read to the active context, so the server checks the workspace-level audit.view permission. Switch workspaces with dome context use <name> before running the command. Tenant- or org-scoped API keys with no workspace selected fall back to the tenant-level audit.view.

Query events

dome audit query

Query audit events with optional filters.

FlagTypeDescription
--typesstring[]Filter by event types
--classesstring[]governing, state_change, security, audit_access
--stagesstring[]attempted, completed
--resultsstring[]allowed, denied
--deny-reasonsstring[]Shared denial reasons such as permission_denied, guard_blocked, or plan_limit
--has-errorboolFilter by error presence. Unset means no filter; --has-error=false selects events without errors.
--act-as-substringSelect governed calls made for one verified end-user subject
--agent-idstringFilter by agent ID
--agent-idsstring[]Filter by agent IDs
--primary-resource-idstringFilter by primary resource ID
--primary-resource-kindstringFilter by primary resource kind
--trace-idstringFilter by trace ID — events for one request
--activity-idstringFilter by activity chain ID. Refer to Activity chains reference.
--operation-idstringFilter by operation ID
--parent-event-idstringFilter by parent event ID
--actor-kindstringFilter by actor kind
--actor-idstringFilter by actor ID
--producer-servicestringFilter by producer service
--producer-gateway-idstringFilter by producer Gateway ID
--request-surfacestringFilter by request surface
--caller-surfacestringFilter by caller application surface
--payload-filterstring[]event.type:field or event.type:field=value (repeatable). Refer to Payload filters reference.
--sincestringStart time (RFC3339)
--untilstringEnd time (RFC3339)
--limitint32Maximum results (default: 50)
--page-tokenstringCursor from a previous query
--compactboolBounded investigation rows; use audit get for full payload

Stream events

dome audit stream

Stream audit events in real-time via SSE. Accepts the same filters as query except --limit and --compact.

FlagTypeDescription
--typesstring[]Filter by event types
--classesstring[]Filter by event classes
--stagesstring[]Filter by operation stage
--resultsstring[]Filter by operation verdict
--deny-reasonsstring[]Filter denied operations by shared denial reason
--has-errorboolFilter by error presence
--agent-idstringFilter by agent ID
--agent-idsstring[]Filter by agent IDs
--primary-resource-idstringFilter by primary resource ID
--primary-resource-kindstringFilter by primary resource kind
--trace-idstringFilter by trace ID
--activity-idstringFollow one activity chain live
--operation-idstringFilter by operation ID
--parent-event-idstringFilter by parent event ID
--actor-kindstringFilter by actor kind
--actor-idstringFilter by actor ID
--producer-servicestringFilter by producer service
--producer-gateway-idstringFilter by producer Gateway ID
--request-surfacestringFilter by request surface
--caller-surfacestringFilter by caller application surface
--payload-filterstring[]Payload filter (repeatable)
--sincestringStart time (RFC3339)
--untilstringEnd time (RFC3339)
--page-tokenstringCursor from a previous stream

Get event

dome audit get [event-id]

Get a single audit event by ID.

Enum values — event class, result, actor kind, surface — render as short tokens (governing, denied, gateway_mcp) in both the table and --format json output, and the filter flags accept those same tokens. Refer to Enum rendering reference.

Export events

dome audit export

Export audit events in JSON-lines, JSON, CSV, OCSF, or CEF format. The command pages to exhaustion up to --limit; it never formats only the first server page. It refuses an artifact when any page is incomplete, repeats a cursor, or would exceed the 10,000-event hard maximum. An unknown payload type is retained with an @unresolved marker and a compatibility warning instead of dropping the event.

FormatOutput
jsonl (default)One JSON object per line (NDJSON).
jsonSingle pretty-printed JSON array.
csvComma-separated rows.
ocsfOCSF v1.1.0 JSON array for Splunk, Sentinel, or Chronicle.
cefCommon Event Format lines for ArcSight, QRadar, or generic syslog.

Accepts the same filters as query, plus:

FlagTypeDescription
--limitint32Maximum results (default: 1000)
--formatstringjsonl, json, csv, ocsf, cef (default: jsonl)
# Export the last 24 hours as OCSF for a SIEM
dome audit export \
  --since 2026-05-26T00:00:00Z \
  --format ocsf > audit.ocsf.json

Catalog

dome audit catalog list

List every audit event type. Complete and unpaginated. Requires audit.view.

dome audit catalog show [event-type]

Show field detail for one event type.

ArgTypeRequiredDescription
event-typestringYesEvent type name

SIEM

Continuous audit → SIEM streaming for Datadog or Grafana Loki. Distinct from dome audit stream (live tail) and from batch export.

Use dome integrations list to find the slug or instance ID of a connected destination.

CommandPermission
dome audit siem listaudit.view
dome audit siem status [slug-or-id]audit.view
dome audit siem enable [slug-or-id]audit.export
dome audit siem disable [slug-or-id]audit.export
dome audit siem resume [slug-or-id]audit.export
dome audit siem list
dome audit siem enable <slug-or-id>
dome audit siem resume <slug-or-id> --skip-to-now --yes

Streaming is forward-only from enable time. enable creates a stream; if one exists, use resume. disable keeps the cursor. --skip-to-now abandons backlog permanently.

enabled and delivery state are separate. A disabled stream retains its last health state, so state=active does not mean it is delivering unless enabled=true. The CLI combines both facts as stopped, stopped (was erroring), or a corresponding paused label. A destination that has never been enabled is absent from list, and status returns not found.

Hash chains

List chains

dome audit chains

List activity chains in the active workspace. A chain is the set of audit events sharing one activity_id — a session, an LLM turn, a conversation thread, or a scripted run — computed at read time by grouping on activity_id within the workspace.

A chain is listed when any of its events matches the filter. Per-chain counts always reflect the full chain. Drill into one chain's events with dome audit chain <activity-id>. Refer to Activity chains reference for the model and trust labels.

FlagTypeDescription
--typesstring[]Chain contains an event of these types
--classesstring[]Chain contains an event of these classes (governing, state_change, security, audit_access)
--stagesstring[]Chain contains an event at one of these stages
--resultsstring[]Chain contains an event with one of these verdicts (allowed, denied)
--deny-reasonsstring[]Chain contains a refusal with one of these reasons
--has-errorboolChain contains an event matching error presence
--agent-idstringChain contains an event for this agent
--agent-idsstring[]Chain contains an event for these agents
--actor-kindstringChain contains an event with this actor kind
--actor-idstringChain contains an event with this actor ID
--producer-servicestringChain contains an event from this producer service
--request-surfacestringChain contains an event from this request surface
--sincestringStart time (RFC3339)
--untilstringEnd time (RFC3339)
--limitint32Maximum chains (default: 50)
--page-tokenstringCursor from a previous chains query

Per-event identity filters (--trace-id, --operation-id, --parent-event-id) and payload filters identify a single event, not a chain, and are not offered here.

# Recent chains in the active workspace
dome audit chains --limit 20

# Chains that contain a denied governing event in the last day
dome audit chains \
  --classes governing \
  --results denied \
  --since 2026-06-05T00:00:00Z

Each row reports the chain's activity_id, event count, last activity timestamp, outcome breakdown, trust label (server-owned, caller-asserted, or mixed — refer to Activity trust reference), and distinct event-type count.

Show chain

dome audit chain [activity-id]

Show one activity chain's events in order. A drill-in into audit query filtered by activity_id. Accepts the same per-event filters as audit query to narrow the chain's events further.

FlagTypeDescription
--typesstring[]Filter by event types
--classesstring[]Filter by event classes
--stagesstring[]Filter by operation stage
--resultsstring[]Filter by operation verdict
--deny-reasonsstring[]Filter by denial reason
--has-errorboolFilter by error presence
--agent-idstringFilter by agent ID
--actor-kindstringFilter by actor kind
--actor-idstringFilter by actor ID
--payload-filterstring[]Payload filter as event.type:field or event.type:field=value (repeatable)
--sincestringStart time (RFC3339)
--untilstringEnd time (RFC3339)
--limitint32Maximum events (default: 200)
--page-tokenstringCursor from a previous query
# Every event in one chain
dome audit chain case-12345

# Only the denied events in that chain
dome audit chain case-12345 --results denied

Emit events

Set --activity (or DOME_ACTIVITY_ID) on the root command to assert a stable activity ID. Every Connect RPC the run makes carries X-Dome-Activity-Id, so the audit events all share one activity_id and form one chain.

# One run, many commands, one chain.
export DOME_ACTIVITY_ID="case-12345"

dome agents register --name worker
dome rules apply ./rules
dome audit query --limit 5

# Inspect the chain you just emitted.
dome audit chains --since 2026-06-05T00:00:00Z
dome audit chain case-12345

--activity overrides DOME_ACTIVITY_ID. The ID is opaque to Dome — pick whatever maps to your unit of work. The header is first-party only and is never forwarded to third-party LLM or MCP backends.

The --activity emit flag (root command) and the --activity-id filter flag (audit reads) are distinct: one stamps events, the other selects them.

On this page

Was this page helpful?