Paste a tool-calling trace. See if the structure is repeating.

The problem

Most teams can see individual tool calls. The harder question is whether the workflow is still moving or just cycling. CAUM turns a run into structural signals: exact cycles, pattern cycles, stalls, burn proxy, and work-yield proxy.

Use CAUM as the first local check before deciding whether a workflow needs a saved receipt or continuous Live monitoring.

Useful event fields

You do not need prompts, source code, documents, customer messages, or private payloads. Structural fields are enough for a first check.

• event type
• tool name
• phase
• status
• state_id/fingerprint
• tokens/cost/latency if available

Example shape

[{"event":"tool_call","tool":"api_lookup","phase":"act","status":"ok","state_id":"lookup_a"},
 {"event":"tool_call","tool":"api_write","phase":"act","status":"error","state_id":"write_error"},
 {"event":"tool_call","tool":"api_lookup","phase":"act","status":"ok","state_id":"lookup_a"}]

What the free check returns

• events analyzed
• exact cycle coverage
• structural pattern cycle coverage
• work-yield proxy
• burn proxy
• trace quality