The seven report types

Driftproof publishes seven kinds of report. Where one compares a model with the skill and without it, it calls a change the same way. They differ in what changes underneath the skill; the value report also counts what a skill costs, and the instrument comparison compares the test tools themselves, so its example has no run without the skill. Each type's name, and each example's title, is the report's own.

Release drift report

Asks whether a skill still helps when a newer version of the model underneath it comes out.

Example: Report 014, Claude Haiku 5.5 on release day, three skills.

Instrument comparison report

Keeps the skills and their tasks the same and measures them with two instruments, two different test tools, saying what each can and cannot show.

Example: Report 010, repeated ADR evaluations across two Claude Code configurations.

Instrument re-measurement report

Changes neither, and runs again tests an earlier report published, with a corrected way of measuring.

Example: Report 007, the baseline is the noisy arm.

Revision drift report

Keeps the model and the tools the same and changes only the skill's own text.

Example: Report 006, the skill text moves, the substrate holds still.

Value report

Keeps the model the same and asks what a skill costs to run, beside whether it helps.

Example: Report 005, what does a skill cost to run?

Capability-gap report

Asks whether a skill still helps on a stronger model from the same company.

Example: Report 004, does encoded expertise still lift output on the frontier tier?

Substrate durability report

Asks whether a skill helps just as much on a different company's model, reached through that company's own command-line tool.

Example: Report 002, does a skill's benefit hold across substrates?