Does Accessibility audit help? Tested on accessibility tasks
Accessibility audit, from rampstackco/claude-skills. Its accessibility audit tasks were run with the skill loaded and without it. Each time it is the same task, run twice.
What we tested
Whether loading this skill helps a model find accessibility problems on a page.
What counts as helping
The skill has to find at least 10 percentage points more of the planted problems than the model finds without it, on the same 10 test pages.
How we scored it
We planted 48 known problems across the 10 test pages and counted how many the model named. The score is the share it found, from 0 to 1.
Without to with, per model
Via API tested via API
Gemini 3.1 Flash Lite First run, with and without the skill
No measured effect
Gemini 3.1 Flash Lite Wider run, three ways
No measured effect
Claude Haiku 4.5 First run, with and without the skill
Holds
GPT-5 mini First run, with and without the skill
No measured effect
GPT-5 mini Wider run, three ways
No measured effect
deterministic pass rate, 0 to 1
In Claude Code tested in Claude Code
Claude Fable 5.1 Fable 5.1, single attempt
No measured effect
Claude Fable 5.1 Wider run, three ways
No measured effect
Claude Fable 5 Second run, every task twice
No measured effect
Claude Opus 5 First run, single attempt
No measured effect
Claude Opus 5 Second run, every task twice
No measured effect
Claude Opus 5 Wider run, three ways
No measured effect
Claude Sonnet 5 Wider run, three ways
No measured effect
Claude Sonnet 5 Second run, every task twice
No measured effect
Claude Haiku 4.5 Second run, every task twice
Holds
Claude Haiku 4.5 Wider run, three ways
No measured effect
Claude Fable 5 First run, single attempt
served model mismatch; 1 usable pair
Withheld
deterministic pass rate, 0 to 1
Show per-task detail
Every item, without and with, per model
Via API tested via API
- Claude Haiku 4.5
- GPT-5 mini
- Gemini 3.1 Flash Lite
- Gemini 3.1 Flash Lite
- GPT-5 mini
deterministic pass rate, 0 to 1
Hover or focus an item to read its task and every model’s two values.
Task-paired: these item means are what this cell's interval is built from.
The item means behind this plot
| Item | Claude Haiku 4.5 | GPT-5 mini | Gemini 3.1 Flash Lite | Gemini 3.1 Flash Lite | GPT-5 mini | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Task | without | with | without | with | without | with | without | with | without | with |
| a10 | 0.250 | 0.500 | 0.250 | 0.500 | 0.250 | 0.750 | 0.250 | 0.750 | 0.250 | 0.500 |
| a06 | 0.400 | 0.600 | 0.400 | 0.400 | 0.400 | 0.600 | 0.400 | 0.600 | 0.400 | 0.400 |
| a08 | 0.000 | 0.667 | 0.333 | 0.667 | 0.667 | 1.000 | 0.667 | 1.000 | 0.667 | 0.667 |
| a03 | 0.400 | 0.600 | 0.600 | 0.400 | 0.200 | 0.200 | 0.600 | 0.200 | 0.600 | 0.600 |
| a05 | 0.600 | 0.600 | 0.000 | 0.000 | 0.800 | 0.600 | 0.800 | 0.600 | 0.800 | 0.600 |
| a01 | 0.500 | 0.875 | 0.625 | 0.875 | 0.750 | 0.875 | 0.750 | 0.875 | 0.625 | 0.875 |
| a09 | 0.333 | 1.000 | 0.667 | 1.000 | 0.667 | 0.667 | 0.667 | 0.667 | 1.000 | 0.667 |
| a04 | 0.200 | 0.400 | 1.000 | 1.000 | 0.600 | 1.000 | 0.600 | 1.000 | 1.000 | 1.000 |
| a02 | 0.667 | 1.000 | 0.667 | 0.667 | 1.000 | 0.833 | 1.000 | 0.833 | 0.833 | 0.833 |
| a07 | 0.750 | 0.750 | 0.750 | 0.750 | 1.000 | 0.750 | 1.000 | 0.750 | 0.750 | 0.750 |
Without to with, per model
In Claude Code tested in Claude Code
Claude Haiku 4.5 Expansion cohort (three arms)
No measured effect
Claude Haiku 4.5 Panel v2 (repeat sampling)
Holds
Claude Fable 5 Panel v2 (repeat sampling)
No measured effect
Claude Sonnet 5 Panel v2 (repeat sampling)
No measured effect
Claude Sonnet 5 Expansion cohort (three arms)
No measured effect
Claude Fable 5.1 Expansion cohort (three arms)
No measured effect
Claude Fable 5.1 Fable 5.1, one pass
No measured effect
Claude Opus 5 Skill cohort (one pass)
No measured effect
Claude Opus 5 Panel v2 (repeat sampling)
No measured effect
Claude Opus 5 Expansion cohort (three arms)
No measured effect
deterministic pass rate, 0 to 1
The Claude Code runs committed their cells as arm means and intervals rather than as per-item values, so this panel is drawn without to with rather than item by item.
What exactly was tested, and how it was scored
- Repository
- rampstackco/claude-skills
- Path
- skills/accessibility-audit
- Commit
0479242522549dfdb389bb9b7807ad4d6016ffb7- Content hash
7b564cd3147768a1e9775a739e55fe56248e996c3fcdab86c264fe7289644921- Date tested
- 2026-08-30
The pin is the whole of this skill's identity here. It resolves at https://github.com/rampstackco/claude-skills/tree/0479242522549dfdb389bb9b7807ad4d6016ffb7/skills/accessibility-audit, and the content hash is a sha256 over exactly the text the model was given with the skill loaded. Nothing else about the skill appears on this site.
S03-accessibility-audit
Loading skills/accessibility-audit from rampstackco/claude-skills improves outputs on accessibility audit tasks.
- Pass criterion
- With-arm mean coverage exceeds the without-arm by at least 0.10 on the same 10 fixture pages at the same model and settings.
- Scale
- unit. Defects found over defects seeded, so a [0,1] fraction, which is the unit scale.
- Task pairs planned per model
- 10
- Notes
- 48 defects seeded across ten pages, ledger at fixtures/accessibility/defect-ledger.json. False positives are counted on the record and never netted off coverage.
Injected context tokens
| Arm | Context characters | Injected context tokens |
|---|---|---|
| without | 0 | 0 |
| with | 43,620 | 10,239 |
The character count is exact: it is the length of the text the with arm is given, and the content hash above is a sha256 over that same text. The token figure is an estimate at 4.26 characters per token, the ratio the phase 1 run measured over 3,120 calls, and it is labelled an estimate until a run reports its own token counts. The without arm is given the identical prompt and nothing else, so its zero is a measurement rather than a missing value.
Verdict per model
| Model | Reading | Orbit | Without -> with | Effect | 95 percent interval | Task pairs | Model version returned |
|---|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | Deterministic, scored against a committed answer key | Stable | 0.4100 -> 0.6992 | +0.2892 | [0.1450, 0.4333] | 10 pairs, 20 of 20 records | claude-haiku-4-5-20251001 |
| Gemini 3.1 Flash Lite | Deterministic, scored against a committed answer key | In free drift unclear. The model scored 63% without it. | 0.6333 -> 0.7275 | +0.0942 | [-0.0683, 0.2566] | 10 pairs, 20 of 20 records | gemini-3.1-flash-lite |
| GPT-5 mini | Deterministic, scored against a committed answer key | In free drift unclear. The model scored 69% without it. | 0.6925 -> 0.6892 | -0.0033 | [-0.1119, 0.1052] | 10 pairs, 20 of 20 records | gpt-5-mini-2025-08-07 |
| Gemini 3.1 Flash Lite | Deterministic, scored against a committed answer key | In free drift unclear. The model scored 63% without it. | 0.6733 -> 0.7275 | +0.0542 | [-0.1349, 0.2433] | 10 pairs, 30 of 30 records | gemini-3.1-flash-lite |
| GPT-5 mini | Deterministic, scored against a committed answer key | In free drift unclear. The model scored 57% without it. | 0.5292 -> 0.6258 | +0.0967 | [-0.0153, 0.2086] | 10 pairs, 30 of 30 records | gpt-5-mini-2025-08-07 |
- Stable: the treatment arm outscored the control arm by more than the threshold, and the interval excludes zero.
- In free drift: no separation the design can resolve. Not evidence of no effect.
Cost
| Model | Arm | Tasks attempted | Mean input tokens | Mean output tokens | Cost per task | Cost ratio | Grading cost per task pair |
|---|---|---|---|---|---|---|---|
Claude Haiku 4.5 served claude-haiku-4-5-20251001 | without | 10 | 600 | 55 | $0.00087 | ||
Claude Haiku 4.5 served claude-haiku-4-5-20251001 | with | 10 | 12,598 | 68 | $0.01294 | 14.83x | |
GPT-5 mini served gpt-5-mini-2025-08-07 | without | 10 | 515 | 115 | $0.00036 | ||
GPT-5 mini served gpt-5-mini-2025-08-07 | with | 10 | 10,995 | 115 | $0.00298 | 8.30x | |
| Gemini 3.1 Flash Lite | without | 10 | 528 | 59 | $0.00022 | ||
| Gemini 3.1 Flash Lite | with | 10 | 11,595 | 67 | $0.00300 | 13.61x |
Defined in the metrics canon. Cost per task divides every dollar spent on an arm by the tasks attempted on it, including tasks whose call returned nothing, because a call that returned nothing was still billed.
Where the with arm did not win
Claude Haiku 4.5, Deterministic, scored against a committed answer key
Drew on 2 of 10 pairs: a05, a07.
Findings not in the answer key: 19 with the skill, 22 without. These are counted and never netted off coverage.
Gemini 3.1 Flash Lite, Deterministic, scored against a committed answer key
Lost on 3 of 10 pairs: a07 (-0.2500), a05 (-0.2000), a02 (-0.1667).
Drew on 2 of 10 pairs: a03, a09.
Findings not in the answer key: 20 with the skill, 18 without. These are counted and never netted off coverage.
GPT-5 mini, Deterministic, scored against a committed answer key
Lost on 2 of 10 pairs: a09 (-0.3333), a05 (-0.2000).
Drew on 6 of 10 pairs: a02, a03, a04, a06, a07, a08.
Findings not in the answer key: 59 with the skill, 64 without. These are counted and never netted off coverage.
Gemini 3.1 Flash Lite, Deterministic, scored against a committed answer key
Lost on 4 of 10 pairs: a03 (-0.4000), a07 (-0.2500), a05 (-0.2000), a02 (-0.1667).
Drew on 1 of 10 pairs: a09.
Findings not in the answer key: 23 with the skill, 18 without. These are counted and never netted off coverage.
GPT-5 mini, Deterministic, scored against a committed answer key
Lost on 1 of 10 pairs: a03 (-0.2000).
Drew on 5 of 10 pairs: a02, a04, a05, a06, a07.
Findings not in the answer key: 67 with the skill, 69 without. These are counted and never netted off coverage.
In Claude Code
These cells are tested in Claude Code, on a subscription path with no API key. They are a second instrument: no figure here is averaged with one tested via API above. How the two were compared.
- Claude Opus 5 · unclear, and both runs agree
- Claude Fable 5 · unclear under repeat sampling
- Claude Haiku 4.5 · helped under repeat sampling
- Claude Sonnet 5 · unclear under repeat sampling
- Claude Fable 5.1 · unclear in the version re-test
Skill cohort (one pass)
one pass over each committed item. Matrix hash 2707a06a. No pre-registration: this run predates the practice on this arm. Run report.
| Model | No skill | With skill | Delta | Interval | Pairs | Orbit | Injected tokens | False positives per page |
|---|---|---|---|---|---|---|---|---|
| Claude Opus 5 | 0.8900 | 0.8700 | -0.020 | -0.143 to 0.103 | 10 | In free drift | 16,699 | 28 with, 15 without |
Verdict withheld on this run
Claude Fable 5: served model mismatch; 1 usable pair. The plan served a different model on most of one arm, so these records are evidence about that other model and are scored into no cell. They are kept, and their tokens still count against what the run consumed.
Panel v2 (repeat sampling)
two passes over each committed item, under repeat sampling. Matrix hash 94bab960. Pre-registration · Run report.
| Model | No skill | With skill | Delta | Interval | Pairs | Orbit | Injected tokens |
|---|---|---|---|---|---|---|---|
| Claude Fable 5 | 0.8000 | 0.8000 | +0.000 | 0.000 to 0.000 | 2 | In free drift | 387,982 |
| Claude Haiku 4.5 | 0.4583 | 0.6904 | +0.232 | 0.088 to 0.377 | 10 | Stable | 270,552 |
| Claude Opus 5 | 0.8900 | 0.8400 | -0.050 | -0.178 to 0.078 | 10 | In free drift | 371,812 |
| Claude Sonnet 5 | 0.8350 | 0.7775 | -0.058 | -0.140 to 0.025 | 10 | In free drift | 384,852 |
| Model | Arm | Records that reasoned | Mean thinking tokens | Most on one record |
|---|---|---|---|---|
| Claude Fable 5 | with skill | 20 of 20 | 1,405 | 3,393 |
| Claude Fable 5 | no skill | 38 of 38 | 522 | 965 |
Fable 5.1, one pass
one pass over each committed item. Matrix hash db373661. Pre-registration · Run report.
| Model | No skill | With skill | Delta | Interval | Pairs | Orbit | Injected tokens | False positives per page |
|---|---|---|---|---|---|---|---|---|
| Claude Fable 5.1 | 0.8750 | 0.9350 | +0.060 | -0.024 to 0.144 | 20 | In free drift | not measured | 12 with, 11 without |
Expansion cohort (three arms)
one pass over each committed item. Matrix hash 0078cfa8. Pre-registration · Run report.
| Model | No skill | With skill | Delta | Interval | Pairs | Orbit | Injected tokens | False positives per page |
|---|---|---|---|---|---|---|---|---|
| Claude Fable 5.1 | 0.8550 | 0.8950 | +0.040 | -0.012 to 0.092 | 10 | In free drift | not measured: this run records the dose per arm, in the doses list | 19 with, 13 without |
| Claude Haiku 4.5 | 0.4550 | 0.6283 | +0.173 | 0.030 to 0.317 | 10 | In free drift | not measured: this run records the dose per arm, in the doses list | 18 with, 20 without |
| Claude Opus 5 | 0.9100 | 0.8300 | -0.080 | -0.200 to 0.040 | 10 | In free drift | not measured: this run records the dose per arm, in the doses list | 29 with, 16 without |
| Claude Sonnet 5 | 0.8375 | 0.8042 | -0.033 | -0.111 to 0.044 | 10 | In free drift | not measured: this run records the dose per arm, in the doses list | 46 with, 37 without |
Every figure above is computed at build time from harness/results/runs-skills.jsonl and harness/results/ledger-skills.jsonl, both committed, by the frozen status_v1 rule and the metrics_v1 cost definitions. How a claim gets tested.