fail
impeccable
It missed 4 of 10 requests it should have handled — the skill stayed silent and the model answered on its own.
Fired when it should have
fired in 6 of 10 attempts
95% confidence between 31% and 83%. 10 attempts leave this very unsettled — run more to narrow it.
Was right when it fired
was right in 6 of 6 attempts
95% confidence between 61% and 100%. 6 attempts narrow it this far; more would narrow it further.
Every case
Instability
2 cases produced both outcomes
- trigger.positive.settings_rework — 1 passed, 1 failed
- trigger.positive.empty_state — 1 passed, 1 failed
Conditions
Two runs are comparable only when all of these are identical. A score that moved because the model changed is not a regression in the skill.
- Skill version
- pbakaus/impeccable@831cabee8b4bc1a2b66e5ae22003e9a19b57d464
- Skill hash
- sha256:5c5260774d0095b797526ecb6f43b01000caf5b70346d07bb89168dcd148cdcb
- Model
- claude-haiku-4-5-20251001
- System prompt hash
- not-provided-by-host
- Environment hash
- sha256:3abcf17d1ef9845506852d41cfa86d509e4cd02c6d7347a69ca58456e672a2d7
- Permission mode
- acceptEdits
- Host memory
- not measured (the record predates 0.4.5 or the probe did not run)
- Context hash
- not measured (the record predates 0.4.5 or the probe did not run)
- Case set version
- 1
- Case set hash
- sha256:c820aafc5cc50f635fb7c46b864eb8ed1642b67bfea9daac2e634c8581a33545
- Assay version
- 0.3.1 or earlier (the record predates version stamping)
Recorded before the host CLAUDE.md leak was closed (Assay 0.4.5). This run’s working directory sat under the measuring machine’s home directory, and Claude Code loads instruction files from every directory above the one it works in, so that machine’s ~/.claude/CLAUDE.md — a single instruction about an unrelated tool, graphify — would have entered the context. Which instruction files actually entered it was not measured.
The name graphify appears in none of this run’s 24 attempts. No trace does not prove no effect.
Not re-measured: no Skill call in any published run targets graphify, and the one finding the file could plausibly affect — the marketing-skills positioning case, 10 of 10 attempts — was reproduced without it, 8 of 10.
No earlier run was measured under the same conditions. See what changed since the previous run
What it cost
- Attempts
- 24
- Tokens in / out
- 1,549 / 73,827
- Cost
- $1.59
- Agent time
- 21 min
Combined trigger score (F1) 0.75 · 20 of 24 trigger decisions were correct.