Observation before claims.

We report what retained evidence supports. Missing values remain open.

How case reports are prepared

1. Select owned documented cases with relevant, readable evidence.
2. Read dated sources, versions and execution paths separately.
3. Remove personal data, access information and private logs.
4. Publish the question, retrospective selection, method, results, failures and limits together.
5. Retain extracts with SHA-256 fingerprints and downloadable notes.

What the labels mean

Observed means a relevant documented observation within the stated check level. Failure observed preserves an actual documented failure. Open means the considered record does not answer the question. Not applicable applies to levels the case does not investigate.

Future study plan: skill comparison

The original FP05 concept asks whether a particular skill revision improves quality or total duration on comparable tasks. No suitable controlled cohort or causal result is presented here. The following steps are a plan, not an executed experiment.

1. Define equally difficult tasks, fixed model/tool conditions and the complete time basis in advance.
2. Control order, retain errors and rework, and run several comparable repetitions.
3. Assess quality independently using the same acceptance criteria.
4. Report raw measurements, missing values and every outcome; do not infer speed from individual structural values.

Provenance and conflicts of interest

Sources are owned project evidence dated 4–6 October 2026. This editorial preparation is dated 7 October 2026. The operator was also involved in the described projects. The cases have not received external scientific peer review. Screenshots show historical owned interfaces, not current service states.

New case IDs KF01–KF04 stay separate from original research concepts FP01–FP06. Downloads contain newly authored notes and redacted extracts; included brands and external project software retain their own rights.