Chat tool telemetry

Find slow, failing, and repetitive agent tools without exposing prompts, arguments, or outputs.

PeriodToolTeam

Detailed production timing active

6035 of 6100 saved tool calls include execution duration.

98.9% timing coverage

Tool calls

+20% vs previous 7d6 saved production turns

p50 latency

350 ms

New activity vs previous 7d6 timed calls

p95 latency

2.6 s

New activity vs previous 7d6 timed calls

Tool failures

0.0%

-100.0 pts vs previous 7d0 excluding NoResults

No result

0.0%

No change vs previous 7d0 product outcomes

Calls / turn

1.0

-80% vs previous 7d33.3% of turns at 5+ calls

What changed

Material regressions ranked by severity and affected call volume.

Current 7 days compared with the previous equal window

Stable against baseline

No material tool regressions detected

Tools with enough activity were checked for higher failure rates, available p95 latency regressions, more no-results, and increased loop pressure.

Review the tool comparison for absolute hotspots that may still need a product threshold.

Investigate

Regression checks require at least 3 calls in both windows. Latency is flagged at +20%; failure and no-result rates are flagged at +1 and +3 points respectively.

Execution volume

Successful calls, no-results, and failures combined.

Calls

Oct 1Oct 4Oct 7

Payload pressure

How often results exceed the 20 KB inline cap.

5.0%

handle rows / all calls

307 stored handles in this period. This is a payload-size proxy, not a one-to-one call classification.

Tool comparison

Each change is measured against the previous equal window; latency appears only where detailed timing exists.

Sorted by current p95 latency

Tool	Calls	p50	p95	Failures	No result
save_attachment_to_workspace	6+20% vs prior	350 msNew activity vs prior	2.6 sNew activity vs prior	0.0%-100.0 pts vs prior	0.0%No change vs prior

Failure codes

Operational failures only; NoResults is excluded.

No operational failures

in the selected view

Interpretation notes

The data has known edges that change what a chart means.

Blocked loops are invisible: Rate-limited and identical-argument calls are rejected before telemetry fires.
NoResults is not downtime: It remains a failed tool outcome, but is separated from runtime reliability here.
Preview volume lives elsewhere: Preview and threadless chats reach PostHog but are intentionally absent from Postgres.
Latency is partially instrumented: Saved tool parts provide production outcomes and volume; duration appears only on calls with detailed telemetry.
Partial turns are retained: They remain in volume counts but are excluded from failures and latency percentiles.
Comparisons use equal windows: Every delta compares the selected period with the immediately preceding period under the same tool and team filters.

Recent sanitized calls

Arguments, outputs, user IDs, and message content are never rendered.

Privacy safe

save_attachment_to_workspaceTexas Materials2.2 sOct 7, 12:39 PM UTC

Outcome

Success

Thread

d64ea9ad…

Tool call

call_Deq…

Turn sequence

1 of 2

Turn state

Completed

save_attachment_to_workspaceTexas Materials326 msOct 7, 11:50 AM UTC

Outcome

Success

Thread

3f067197…

Tool call

call_6Yr…

Turn sequence

1 of 2

Turn state

Completed

save_attachment_to_workspaceTexas Materials2.2 sOct 7, 11:38 AM UTC

Outcome

Success

Thread

3f067197…

Tool call

call_e0k…

Turn sequence

1 of 2

Turn state

Completed

save_attachment_to_workspaceTRP 2350 msSep 30, 3:45 PM UTC

Outcome

Success

Thread

a8fe7c6c…

Tool call

call_cpR…

Turn sequence

1 of 2

Turn state

Completed

save_attachment_to_workspaceTRP 2264 msSep 30, 3:23 PM UTC

Outcome

Success

Thread

dfa4cc36…

Tool call

call_Npj…

Turn sequence

16 of 29

Turn state

Completed

save_attachment_to_workspaceTRP 22.6 sSep 30, 2:44 PM UTC

Outcome

Success

Thread

a7197a7d…

Tool call

call_307…

Turn sequence

3 of 14

Turn state

Completed