AB icp-builder
B2B SaaS ICP (Ideal Customer Profile) and buyer persona builder. Use this skill whenever the user wants to define, refine, or document their ideal customer profile, buyer personas, target segments, pain points, or jobs-to-be-done (JTBD). Trigger for phrases like "who is my ICP", "define my target customer", "build a persona", "who should we target", "what are our buyers pain points", "jobs to be done", "customer segments", "who are we selling to", "ideal customer", "target audience for B2B", "firmographics", or any request to understand, document, or sharpen who the company's best customers are. Also trigger when the user is starting go-to-market planning, repositioning, or says they're not sure who to focus on. This skill produces a complete, structured ICP document ready for use in sales, marketing, and product decisions.
As a process B 69/100 · Nearly there — weak spots: inputs and preconditions, consistency, progress reporting
How to improve
- 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: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 69/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 40Consistency. Frontmatter name (icp-builder) differs from the folder (b2b-saas-icp-builder)
- 55Failures and branches. 1 branches
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. No external tools needed
- 100Steps. 8 steps
- 100Execution cost. Instruction body is 809 tokens
- 100Running it twice. No mutating operations
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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 835: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 11 example trigger phrases
- +4Structure: 8 headings
- +3Step-by-step instructions: 8 items
- +3Output format is stated explicitly
- +4Has examples (6 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 89.