AB protocol-deviation-classifier
Determine whether an incident in a clinical trial is a "major deviation.
As a process B 76/100 · Nearly there — weak spots: consistency, running it twice
How to improve
For the model run — optional
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "skill-author"
Process rating: all ten parameters 76/100
- 30Running it twice. 1 mutating operations with no state check
- 40Consistency. Frontmatter name (protocol-deviation-classifier) differs from the folder (protocol-deviation-classifier-1)
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 106 steps
- 100Result and completion. Output format and completion criterion are stated
- 100Failures and branches. 6 branches, has a failure section
- 100Execution cost. Instruction body is 3619 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 26 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 72: 120–800 characters recommended
- -41 reference files, but SKILL.md never points to them: the model will not open them
- +2Single-language instructions
- +4Structure: 35 headings
- +3Step-by-step instructions: 106 items
- +3Output format is stated explicitly
- +4Has examples (8 code blocks)
- +3All 1 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 83.
External checks
ClawHub: clean
This is a local clinical-trial deviation classifier with no evidence of data theft or hidden system access, but its high-stakes recommendations should be reviewed by qualified staff.
LLM: benign (high) · VirusTotal: · 29 May 2026