Trace analysis, answered
TraceIQ ingests a manuscript trace, reconstructs the full call tree, and lets you ask — in plain English — what went wrong. Every answer is grounded in the actual trace, with the evidence attached.
How it works
No dashboards to configure. Bring a trace, get an answer you can act on.
01
It’s parsed end to end — the full call tree, requests, and field values, reconstructed and indexed.
02
“Which rule zeroed out this coverage?” “Why did this field come back empty?” Plain English, no query language.
03
It points at the exact node and shows the trace data behind the conclusion — so you can verify, not just trust.
What it does
The complete tree, every node — never silently truncated, so the analysis reasons over the whole run.
Follow how a value was derived across the execution, step by step, to the point it went wrong.
See the calls behind the behavior — inputs, outputs, and where a response diverged from expectation.
Every claim ties back to a specific node in your trace. Grounded in your data — it reads, it doesn’t guess.
Your data
Your traces are visible only to you. Nothing is shared across accounts.
Answers come from your uploaded trace — the model reads the full tree rather than inventing.
Every trace is retained for 7 days, then permanently deleted — raw file and parsed data alike. Remove any trace sooner whenever you like.
Create an account, upload a trace, and ask what went wrong. Free to start.
Start free