Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.crewship.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
crewship journal

Read the Crew Journal — the canonical append-only event stream for every observable action in the platform. Filter by crew, agent, mission, entry type, severity, actor, priority, trace, or time window. 
crewship journal [flags]
crewship journal get <entry-id>
crewship journal count [flags]
crewship journal priority <entry-id> --mark <marker> [--reason <text>]
The top-level command is the list view. --follow opens an SSE connection to /api/v1/journal/stream and prints entries as they arrive (Ctrl-C to exit). The CLI auto-reconnects on transient failure with bounded exponential backoff (1 s → 30 s) and threads Last-Event-ID so a brief disconnect resumes without dropping or duplicating entries. For pretty in-browser viewing, the web UI /journal page is also available.

​
Flags

FlagTypeDefaultDescription
--linesint50Max entries to fetch (1-500).
--crewstringFilter by crew slug or ID.
--agentstringFilter by agent ID.
--missionstringFilter by mission ID.
--trace-idstringNarrow to a single run’s spans (trace_id == run id).
--typestringComma-separated entry types (e.g. peer.escalation,keeper.decision).
--exclude-typestringComma-separated entry types to exclude (NOT IN). Useful for hiding container.metrics noise.
--severitystringComma-separated severities (info, notice, warn, error).
--actor-typestringComma-separated actors (agent, user, system, keeper, sidecar, orchestrator).
--prioritystringComma-separated priorities (normal, high, pin, permanent).
--query / -qstringServer-side FTS5 search across summary + payload. Bounded to 200 chars.
--sincestringTime window (1h, 24h, 7d, or RFC3339).
--followboolfalseLive tail via SSE — Ctrl-C to exit, reconnects on transient failure.
Global output flags apply: --format text|json|yaml (default text). Enum values for --severity, --actor-type, and --priority are validated client-side before the request goes out, so a typo fails fast with a list of allowed values.

​
Examples

# Last 50 entries across the workspace
crewship journal

# Last 24h of a specific crew
crewship journal --crew backend-team --since 24h

# Every warn/error
crewship journal --severity warn,error

# Recent escalations and keeper decisions
crewship journal --type peer.escalation,keeper.decision --lines 100

# Mission timeline as JSON for scripting
crewship journal --mission MIS-42 --format json

# Full-text search across summary + payload
crewship journal --query "OOMKilled" --since 24h
crewship journal -q "ratelimit" --severity warn,error

# One run's full trace, including LLM/exec/network spans
crewship journal --trace-id <run-id>

# Hide container.metrics noise from a long timeline
crewship journal --exclude-type container.metrics,exec.output_chunk --lines 200

# Only operator-pinned entries
crewship journal --priority pin,permanent

# Live tail — print entries as they arrive (Ctrl-C to exit)
crewship journal --follow
crewship journal --follow --severity warn,error
crewship journal --follow --crew backend-team
For free-text “did we discuss this before?” lookups across the journal, prefer crewship recall <query> — same backing endpoint, snippet-card rendering optimised for question-answering.

​
crewship journal get

Fetch a single entry by ID. Output respects --format (text, json, yaml); the text format prints the same single-line summary the list view uses, then indents the entry’s payload and refs underneath so you can see the full structured detail at a glance. 
crewship journal get j_a1b2c3d4
crewship journal get j_a1b2c3d4 --format json | jq .payload
The endpoint is workspace-scoped — IDs from another workspace come back as 404 with the same shape as “not found” (no cross-tenant existence leak).

​
crewship journal count

Print the total number of entries that match a filter set. Identical filter flags to the list view (minus --lines / --follow); --cursor and --limit are silently ignored — count is always over the full result set. 
crewship journal count
crewship journal count --severity error --since 24h
crewship journal count --type budget.exceeded
crewship journal count --query "OOMKilled" --format json
Default text output is the bare integer for shell scripting: 
if [ "$(crewship journal count --type budget.exceeded --since 1h)" -gt 0 ]; then
  pagerduty-trigger "Budget exceeded in last hour"
fi

​
crewship journal priority

Annotate one entry with a priority marker (normal, high, pin, permanent). Requires OWNER or ADMIN on the workspace. Marker affects compaction and recall — see the Crew Journal guide for the semantics. 
crewship journal priority j_abc --mark permanent --reason "FX compliance rule"
crewship journal priority j_abc --mark pin --reason "team playbook entry"
crewship journal priority j_abc --mark normal
The --mark value is validated client-side before the request goes out. Each invocation also writes a memory.priority_changed audit entry to the journal carrying the previous and new priority plus the supplied reason — so the curation history is itself queryable.

​
Sample output

2026-04-17 10:23:41  [info    ]  peer.conversation       agent       viktor → eva: "pushing hotfix"
2026-04-17 10:24:02  [warn    ]  keeper.decision         keeper      denied production SSH (risk=8)
2026-04-17 10:24:15  [error   ]  budget.exceeded         system      crew/day at 103.2% ($51.62 / $50.00)
Columns: timestamp, severity chip (color-coded — gray=info/default, cyan=notice, yellow=warn, red=error), entry type (bold), actor type, summary.

​
Entry type catalog

See the Crew Journal guide for the full list. Common filters:
--type valueWhen to use
keeper.decisionEvery credential access decision.
peer.escalationCross-agent handoffs (often tied to approval.requested).
budget.exceeded,budget.warningCost oversight.
approval.requested,approval.granted,approval.denied,approval.timeoutHITL audit.
hook.fired,hook.blockedLifecycle hook activity.
eval.regression_detectedEval surfaced regression.
run.started,run.completed,run.failed,run.cancelled,run.timeoutRun lifecycle (/runs page is folded into /journal since PR #234).

​
Parsing --since / --until

Accepted formats:
  • Duration suffixes: 30m, 1h, 24h, 7d (the d suffix is handled locally; Go’s time.ParseDuration doesn’t support it).
  • RFC3339: 2026-04-17T00:00:00Z.
--until is supported on journal count (lower bound is --since, upper bound is --until); the list view doesn’t currently expose --until because the default ordering is “newest first” and pagination uses a cursor.

​
Notes

  • Output is scoped to the caller’s workspace automatically — there is no way to query another workspace from the CLI.
  • Pagination is keyset-based on the server; the list view doesn’t expose the cursor (it fetches --lines rows and stops). For paged programmatic access, prefer the JSON output of the /api/v1/journal endpoint.
  • For structured programmatic access prefer --format json.