Spec your practice. Count every mistake.

All-time, custom ranges, day / week / month — CE, WA, TLE, attempts to AC, and streaks that survive a reinstall.

FIG 01 · ERROR SURFACE

Track mistakes, not just greens

Compilation errors, wrong answers, timeouts — counted with solves. Optional extension attaches compiler text when you open a failed submission.

CE / RE capture

Message-level

Scraped from CF submission pages via the Chrome extension.

Verdict mix

WA · TLE · MLE

Full breakdown per day, week, and month period.

Attempts → AC

2.4

Average tries before first accepted solution.

FIG 02 · PERIOD DELTA

All-time overview, custom From→To ranges, or classic day / week / month deltas — pick the window that matches how you practice.

compare · week vs prevunified
period: 2026-W29
solved: 12   (+3 vs prev)
errors: 9   (−4 vs prev)
streak: 5 days
filter: practice | contest | all

FIG 03 · PERSISTENCE

Neon Postgres holds your ledger. New laptop, reinstall the extension, sign back in — history is still there.

Source of truth

Submissions upsert by id. Reinstall never forks your stats.

Ownership

CF handle + password, with profile-token verification before sync.

FIG 04 · PIPELINE

01
Sign up with your Codeforces handle and a password, then prove the handle is yours.
02
We import submissions in the background — open the dashboard while it loads.
03
Optional: install the extension to capture compiler text from failed CF pages.

FIG 05 · OPTIONAL MODULE

Dashboard works after signup. Install only when you want compiler/runtime messages attached to the log.

$curl -LO /downloads/solvetrace-extension.zipInstall guide

FIG 06 · GLOSSARY

Codeforces verdicts and SolveTrace metrics, in plain language — so the dashboard isn’t a wall of abbreviations.

solved: Distinct problems you got Accepted on in that period. Same problem AC’d three times still counts as one.
OK subs: Total Accepted submissions — every green click. Can be higher than solved if you re-submit ACs.
submissions: Everything you sent to the judge in that period (passes and fails), excluding “in queue / testing”.
errors: Submissions that were not Accepted — WA, CE, TLE, and the rest, counted together.
error rate: Share of submissions that failed: errors ÷ submissions. Lower usually means cleaner practice.
WA: Wrong Answer — program ran, but output didn’t match the expected answer (logic / edge cases).
CE: Compilation Error — code didn’t build. Syntax, missing imports, wrong language settings.
TLE: Time Limit Exceeded — too slow. Need a faster approach or tighter constants.
MLE: Memory Limit Exceeded — used more RAM than allowed. Big arrays, leaks, heavy structures.
RE: Runtime Error — crashed while running (null, overflow, out-of-bounds, stack overflow, etc.).
other: Less common outcomes (idleness limit, rejected, challenged, …) bundled so the chart stays readable.