SKILLEMALL.ai

CF paid-ads-setup

Connect NotFair paid-ad accounts and establish reusable campaign context. Use when setting up paid media in a new workspace, connecting Google, Meta, X, LinkedIn, Reddit, or TikTok Ads, refreshing brand context, or preparing an agent to work on ads safely.

nowork-studio/notfair-plugin Claude Code author: nowork-studio MIT 1 file body ≈ 157 tokens Open the sourcegithub.com↗ analyzed 2 d ago

Connect NotFair paid-ad accounts and establish reusable campaign context.

As a process F 20/100 · No process to follow — References files that are not bundled: ../../paid-ads/paid-ads-setup/WORKFLOW.md

ReferenceMarketingtype and topics are labelled automatically from the skill text
JSON
Technical rating
C
87/100
safety, quality, tests
Safety 60%
100
Quality 40%
67
Run on models
none yet
Process rating
F
20/100
No process to follow
References files that are not bundled: ../../paid-ads/paid-ads-setup/WORKFLOW.md
Tools and files w 18
0
Steps w 15
0
Result and completion w 14
0
the three weakest of ten parameters · all ten

How to improve

  1. The text references files that are not there: add them or drop the references.
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: 1. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning missing-ref reference to a missing file: ../../paid-ads/paid-ads-setup/WORKFLOW.md

Process rating: all ten parameters 20/100

Will not run. References files that are not bundled: ../../paid-ads/paid-ads-setup/WORKFLOW.md
  • 0Tools and files. 1 referenced file(s) missing: ../../paid-ads/paid-ads-setup/WORKFLOW.md
  • 0Steps. Prose only: no discrete steps
  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Failures and branches. Linear process with no failure handling
  • 0Progress reporting. Says nothing while it works
  • 20When it triggers. No condition that starts the skill
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 157 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

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +4Structure: 1 headings, hard to scan
  • +3No numbered steps or checklist
  • +3Output format is not stated: the model decides each time
  • +4No input/output examples
  • +1No license
  • +2Single-language instructions
  • +3Description length 256: enough signal without eating the budget

Quality base 70; lint remarks subtract, signals add up to 100. Result: 67.