Sample report. Every name and number is invented.
Bethan Lloyd-Pryce
PassedLicenceAssessed work at Fernbrook Systems, Q3 2026 so far. Personal training is not counted here and never reaches a team report.
- Submissions
- 5
- Findings caught
- 11 of 12
- Catch rate
- 92%
- Median score
- 100 against a pass mark of 80
Against a frontier model
How a frontier model scored on the same challenges, given the same changes and marked against the same correct answers, with no hints and no walkthrough. The model is not named: these figures are invented for this sample.
- This person
- 100 of 100
- The model, on the same challenges
- 71 of 100
- Difference
- +29 points
Both figures are out of 100 over the 5 challenges of theirs that a published run covers, so they are only compared on work the model has actually done.
Score over time
Every attempt in the order it was made, oldest on the left, against the pass mark.
What they catch and what they miss
The best and worst of this person's own results by weakness category. It is their own detail, not an aggregate, so no minimum cohort applies. The whole breakdown is in the table below.
Strongest
- AI safety100% caught2 found of 20 missed
Weakest
- Injection88% caught7 found of 81 missed
By CWE category
What this reviewer catches and misses, by weakness family.
| Category | Caught | Missed | Catch rate |
|---|---|---|---|
| Injection | 7 | 1 | 88% |
| Data exposure | 2 | 0 | 100% |
| AI safety | 2 | 0 | 100% |
By language
The same breakdown by the language of the codebase reviewed.
| Language | Caught | Missed | Catch rate |
|---|---|---|---|
| PHP | 6 | 1 | 86% |
| Python | 5 | 0 | 100% |
Every attempt
Newest first. Times are in UTC.
| Challenge | Score | Grade | Caught | Took | Submitted |
|---|---|---|---|---|---|
Search results assembled into the prompt deskmindchallenge 2 | 100 | Passed | 3/3 | 30 min | 24 Jul 2026, 10:07 UTC |
A ticket summary that follows the instructions in the ticket deskmindchallenge 1 | 100 | Passed | 2/2 | 27 min | 23 Jul 2026, 04:07 UTC |
An export carrying columns the page never showed marksheetchallenge 3 | 95 | Passed | 2/2 | 25 min | 21 Jul 2026, 22:07 UTC |
A search filter pasted into the where clause marksheetchallenge 2 | 100 | Passed | 2/2 | 23 min | 20 Jul 2026, 16:07 UTC |
A grade query built by concatenation marksheetchallenge 1 | 63 | Needs work | 2/3 | 21 min | 19 Jul 2026, 10:07 UTC |