Driftproof
Menu
Skills.sh snapshot 2026-10-17: text tasks
Worth installing?

media-use

Here is what happened when we tested its written answers, with and without the skill.
About this test

media-use: Skills.sh snapshot 2026-10-17: text tasks by Driftproof; Haiku 5.5 judge; eval account checked before/after runs, not during

Report 015: Skills.sh snapshot 2026-10-17: text tasks

Skill skills/media-use in heygen-com/hyperframes.

Each result has its own test record. Results are not combined.

Measured results

Both model sets use Haiku 5.5 as judge.

Driftproof measured

Answering model: Opus 5.5. Judge: Haiku 5.5. Date: 2026-10-18.

Three text-only tasks: Clearly helped.

At least one task improved, with no task showing a clear drop.

Read test record

Why this result?

The skill clearly helped: 0.909 with it and 0.759 without, on 3 test tasks.

reuse: higher scores with the skill. Compare task scores

Task score bands: reuse (separated-up): with skill 0.77 plus or minus 0.144049; without skill 0.43 plus or minus 0.09083.

A band is the average score plus or minus the recorded spread. These task bands, not the overall averages, determine the result.

0.909 with the skill and 0.759 without; on at least one of the 3 test tasks, the two ranges of scores (the average plus or minus one spread) did not overlap, and the gap was at least 0.05.

Test details and sources
How this result was decided

The reader applies the task rule, with a separated-down task taking precedence. Overall means do not determine the verdict. Bands show the mean plus or minus the recorded spread; point bands can reflect score quantization.

  • reuse: separated-up. With skill 0.77 plus or minus 0.144049; without skill 0.43 plus or minus 0.09083.
  • logo: underpowered. With skill 0.983333 plus or minus 0.028868; without skill 0.958333 plus or minus 0.080104.
  • grade: underpowered. With skill 0.975 plus or minus 0.05; without skill 0.89 plus or minus 0.082158.

Three text-only tasks. Driftproof wrote the suite from the pinned skill description and README. These are not author-supplied tests. External tools, generated media and deployed services were not tested.

The eval account's configuration was read and found clean before and after each run, not during it.

Read receipt · Receipt JSON · Suite · Retained answers

Earlier attempts
Receipt metadata

Attempted generations: 28. Measured generations: 28. Lost draws: 0. Judge samples: 140.

Model
claude-opus-5-5
Judge
claude-haiku-5-5
Repository
heygen-com/hyperframes
Skill path
skills/media-use/SKILL.md
Pinned commit
c4440dfb200def5e0c0afee938778f05b5985039
Harness
claude-code 2.1.287
Receipt hash
ceb928044061eeaec1f43472344257829307fb4785f57a754ca5a7c1937d35f1
Plan hash
af83e8024b571620361e0e1ee1e371c21589e6788f25bbe53542fd11f941da53

Driftproof measured

Answering model: Sonnet 5.5. Judge: Haiku 5.5. Date: 2026-10-18.

Three text-only tasks: Clearly helped.

At least one task improved, with no task showing a clear drop.

Read test record

Why this result?

The skill clearly helped: 0.936 with it and 0.802 without, on 3 test tasks.

reuse, grade: higher scores with the skill. Compare task scores

Task score bands: reuse (separated-up): with skill 0.85 plus or minus 0.05; without skill 0.54 plus or minus 0.082158; grade (separated-up): with skill 0.983333 plus or minus 0.028868; without skill 0.9 plus or minus 0.05.

A band is the average score plus or minus the recorded spread. These task bands, not the overall averages, determine the result.

0.936 with the skill and 0.802 without; on at least one of the 3 test tasks, the two ranges of scores (the average plus or minus one spread) did not overlap, and the gap was at least 0.05.

Test details and sources
How this result was decided

The reader applies the task rule, with a separated-down task taking precedence. Overall means do not determine the verdict. Bands show the mean plus or minus the recorded spread; point bands can reflect score quantization.

  • reuse: separated-up. With skill 0.85 plus or minus 0.05; without skill 0.54 plus or minus 0.082158.
  • logo: underpowered. With skill 0.975 plus or minus 0.05; without skill 0.966667 plus or minus 0.028868.
  • grade: separated-up. With skill 0.983333 plus or minus 0.028868; without skill 0.9 plus or minus 0.05.

Three text-only tasks. Driftproof wrote the suite from the pinned skill description and README. These are not author-supplied tests. External tools, generated media and deployed services were not tested.

The eval account's configuration was read and found clean before and after each run, not during it.

Read receipt · Receipt JSON · Suite · Retained answers

Earlier attempts
Receipt metadata

Attempted generations: 21. Measured generations: 21. Lost draws: 0. Judge samples: 105.

Model
claude-sonnet-5-5
Judge
claude-haiku-5-5
Repository
heygen-com/hyperframes
Skill path
skills/media-use/SKILL.md
Pinned commit
c4440dfb200def5e0c0afee938778f05b5985039
Harness
claude-code 2.1.287
Receipt hash
65231f29050580d7e14324c7aff5bfc053844d5c11093958713393e98283c26c
Plan hash
af83e8024b571620361e0e1ee1e371c21589e6788f25bbe53542fd11f941da53

How we test a skill

A skill gives an AI extra instructions. We ask: do those instructions improve its written answers?

  1. The task Same question. Same AI.

    Driftproof writes tasks from the skill description.

    Give the AI a task
  2. The comparison Change the instructions
    With the skill Written answer
    Without the skill Written answer

    Repeat both ways to see how much answers vary.

  3. The result Judge each task
    Haiku 5.5 scores the answers

    A clear drop on any task takes priority. Otherwise, a better task can mean the skill helped. When answers vary too much, we cannot settle the result.

These tests cover written answers, not installation, tool use or how useful a skill is in general.

Receipts by Driftproof (driftproofhq/driftproof), licensed Apache-2.0.

Disagree, or want a re-run?

Built on from committed receipts.