AgentOps MonitorStart free

Black Friday

50% off for 3 paid months.

New customers on a monthly plan. Then the regular price. Cancel anytime. A Team trial started now is still 14 days free, and the 50% off applies to the first 3 paid months after that. Code BLACKFRIDAY50 is applied at checkout.

Ends December 1, 2026 at 11:59 p.m. Pacific.

Example trace

Find the $0.164 call behind about $400 a night.

AgentOps Monitor records every tool call, LLM call, and dollar your agent reports to it, so an on-call engineer at 2 a.m. can stop guessing and start fixing.

Start monitoring Sign in

Already have an account? Generate a key on the API keys page.

Example · 02:14:11Z · run r_9f2c1a4e · refund-triage-v3Run cost $0.190
Event
6 of 8
Type
llm
Model / tool
openai / gpt-4o
Tokens
64,118 tok
Duration
7,394 ms
Cost
$0.164
Status
ok

Found it. After three retries the agent sent a 64,118-token context back to the model. This one call is $0.164 of the run's $0.190 (86%). At 2,400 runs a night, that is about $394 by morning.

Trace your own agent

Eight timeline events are six grouped span rows plus run start/end markers. The six rows represent eight recorded spans because three Stripe retry attempts share one row. HELD is a status reported by your agent at ingest; AgentOps Monitor records it and does not hold the call.

Same run, two views

Which step cost the most? Try answering from the log.

Your logs today14 lines, 0 dollar figures, 0 token counts
02:14:07.100 INFO  refund-triage-v3 starting job=rf_88213
02:14:07.112 INFO  openai request model=gpt-4o messages=14
02:14:07.712 INFO  openai response 200
02:14:07.920 INFO  vectorstore hits=6
02:14:07.920 INFO  stripe.refunds.create sent
02:14:12.502 WARN  stripe.refunds.create pending, retry 1/3
02:14:14.371 WARN  stripe.refunds.create pending, retry 2/3
02:14:16.240 WARN  stripe.refunds.create pending, retry 3/3
02:14:18.108 INFO  openai request model=gpt-4o messages=41
02:14:18.124 DEBUG prompt="You are a refund triage agent..."
02:14:25.502 INFO  openai response 200
02:14:25.502 INFO  slack.postMessage channel=#refunds
02:14:55.319 INFO  slack ok
02:14:55.319 ERROR job rf_88213 failed: refund not confirmed
The same run in AgentOps MonitorRecorded spans and costs, with retries grouped
Duration
48.2 s
Spans
8
Tokens
72,322
Cost
$0.190
  1. llmgpt-4ook · 600 ms$0.026
  2. toolvectorstore.queryok · 208 ms$0.000
  3. toolstripe.refunds.createheld · 4,582 ms$0.000
  4. toolstripe.refunds.createretried 3 times · 5,606 ms$0.000
  5. llmgpt-4ook · 7,394 ms$0.164
  6. toolslack.postMessageok · 29,817 ms$0.000

Read down the cost column: the second gpt-4o call is $0.164 of the $0.190 run (86%). The log on the left never mentions a dollar. Example values.

8 recorded spans become 6 grouped rows because three Stripe retry attempts share one row. The hero adds run start/end markers to those rows, making 8 timeline events. A status like held is reported by your agent at ingest; AgentOps Monitor records it and does not hold the call.

Start free See the ingest format

How it works

From signup to your first trace in three steps.

Create a key

Sign up free, open the API keys page, and create a key. Keys start with aom_live_ and are shown once. Free includes 10,000 runs a month.

Start free →

Send your first trace

curl -X POST https://agentopsmonitor.com/api/ingest \
  -H "Authorization: Bearer aom_live_..." \
  -H "Content-Type: application/json" \
  -d '{"external_id":"run-001","agent_name":"refund-triage","tokens_in":8204,"tokens_out":512,"cost_usd":0.08,"spans":[{"span_type":"llm","provider":"openai","model":"gpt-4o","tokens_in":8204,"tokens_out":512,"cost_usd":0.08,"duration_ms":600}]}'

A plain HTTPS endpoint. Post from Python, Node, curl, or any HTTP client. There is no SDK to install.

Read the docs →

Read the run list

Every run with its status, tokens, and USD cost, newest first. Open one to see its spans. Red marks a failed run.

AgentTokensCost
refund-triage-v3 (failed)72,322$0.190
support-bot (completed)1,512$0.006
refund-triage-v3 (completed)8,716$0.026
lead-enricher (completed)3,240$0.011

Example values.

Pricing

Free to start. Priced by runs, not seats.

Free up to 10,000 runs a month. Team is $299 with 500,000 runs, after a 14-day free trial. Enterprise starts at $2,000.

Start free Full pricing

Find the $0.164 call behind about $400 a night · AgentOps Monitor