AC tam-building
Stand up your account universe as a deployed pipeline: an AI Ark company search shaped by your ICP, sized for free before it bills, then tiered A / B / C / disqualified by an agent that reads your rubric from the workspace context and web-searches the evidence the sourced row does not carry. Triggers: "our TAM is a stale CSV", "build our account universe", "source companies matching our ICP and rank them", "keep our market list current", "which of these companies are actually worth a rep", "tier the market we just sourced". Cargo CDK, aiArk, countCompanies, fetchCompanies, agent tiering, workspace context, webSearch. Skip when: you want the list once rather than a pipeline that keeps producing it, which is build-tam-list; or the accounts already exist in a CRM or an accounts model and only need judging, which is account-scoring.
Stand up your account universe as a deployed pipeline: an AI Ark company search shaped by your ICP, sized for free before it bills, then tiered A / B / C /…
As a process C 59/100 · Has gaps — weak spots: result and completion, inputs and preconditions, progress reporting
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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: 14. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5603 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 59/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 40Result and completion. Does not say what the result is
- 50When it triggers. No condition that starts the skill
- 55Failures and branches. 1 branches
- 60Tools and files. Uses tools (bash, web, node) that frontmatter does not declare
- 70Execution cost. Instruction body is 5603 tokens
- 100Steps. 30 steps
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
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 840: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 6 example trigger phrases
- +4Structure: 10 headings
- +3Step-by-step instructions: 30 items
- +4Has examples (2 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.