Analytics
Read what the campaign agents earned — per project and per agent, on the dashboard's own windows — and close the loop on the goal.
Analytics closes the Moab loop: read what the campaign agents did, compute what it earned, move the goal, and record the belief that should change the next run.
Project performance
curl --fail-with-body "https://moabcrm.com/api/v1/performance?metric=links_clicked&bucket=day" \
-H "Authorization: Bearer $MOAB_API_KEY" \
-H "Project-ID: $PROJECT_ID"metric is any of the dashboard's measures — links_clicked (default),
cost_usd, replies, opens — and bucket is day or total. The answer
carries the same windows and evidence the product's Performance block draws,
so a number here is the number on the screen. The per-agent read is on
Campaign Agents.
Read the runs you are measuring
Bounded and newest-first — both of these resources grow for the life of the project, so read only the window you are measuring.
List agent runs — GET /api/v1/campaign-agent-runs
curl "https://moabcrm.com/api/v1/campaign-agent-runs" \
-H "Authorization: Bearer $DASH_API_KEY" \
-H "Project-ID: pr_zest"Show me the agent runs filtered by agent iddata_query {
"collection": "agent_runs"
}{
"object": "list",
"data": [
{
"object": "campaign-agent-run",
"id": "obj_123",
"status": "status",
"run_type": "run type",
"summary": "summary",
"details": "details",
"created_at": "2026-07-25T12:00:00.000Z",
"updated_at": "2026-07-25T12:00:00.000Z"
}
],
"has_more": false,
"next_cursor": null
}The resources
analytics— read-only. Measured outcomes per period.campaign-agent-runs— read-only. What each run did.dashboards— read-only. Metric boards built from your CRM data.learnings— read-only over REST. One belief the agent holds about a subject.
Worth knowing
- From outside, write analytics as a goal update plus a plain timeline
sentence — never as structured
analyticsrows. Those belong to theanalyticsskill, which computes them as a whole cycle. The goal update is on Products & Goals. - Learnings are discrete facts, created and removed — no decaying score.
Read them with
learnings_context, add one withlearnings_create. - Nothing here is gated. The approval brake blocks spend, not truth: reads, timeline writes, reporting and goal updates always go through. A paused project still gets its full report.
Let the agent do it
For a number you already hold, do not delegate — write it, then append the plain sentence to the goal's timeline:
timeline_append { "subject_type": "goals", "subject_id": "goal-…",
"kind": "analytics_refresh",
"title": "Signups reached 128 of 200 this week",
"body": "Signups reached 128 of 200 this week, up 14 from last." }Delegate when the work needs Moab's opinion — the full cycle computation:
tasks_create { "project_id": "pr_zest", "skill_id": "analytics", "run": true,
"input": { "goalId": "goal-…" } }Next: Reach your business goal.