CB acceptance-checks
Turns the acceptance criteria a task states into executable checks and runs them against the finished outputs before delivering, rather than trusting that a correct-looking analysis satisfies them; rehearses the hidden condition when the grader scores unseen instances from a described family; treats a stated formula, update rule or predicate as the oracle to implement first; so that every stated path, schema, unit, rounding rule, ordering, tolerance, threshold, named method and version, prescribed heading and self-consistency formula becomes a small test, any checker the task ships is run as the exit criterion, derived quantities are recomputed from the primitives, and conventions such as units, signs, coordinate frames and code tables are transcribed into named functions with hand-checked cases. Use whenever a task states how the result will be judged mechanically, ships a validation script, or specifies output formats and tolerances; a report a reader grades needs only its headings and files checked once, and free-form exploration with no stated criteria does not need it.
Turns the acceptance criteria a task states into executable checks and runs them against the finished outputs before delivering, rather than trusting that a…
As a process B 72/100 · Nearly there — weak spots: result and completion, inputs and preconditions, running it twice
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- Shorten the description to 1024 characters.
- 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 · 1
✓ No critical or high findings
Medium and low: 1
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Read Bash python Grep Glob
Files scanned: 1. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1090 chars, limit 1024 - note
frontmatter-keyunknown frontmatter key "summary" - note
frontmatter-keyunknown frontmatter key "role"
Process rating: all ten parameters 72/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 5 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 4209 tokens
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 32 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
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 1090: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +4Structure: 5 headings
- +3Step-by-step instructions: 32 items
- +4Has examples (0 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 55.