AC find-marketing-agency
Use whenever the user wants to find, shortlist, vet, or enrich US marketing agencies — including branding, content marketing, PPC/paid media, social media, email marketing, performance/demand-gen, video production, and full-service digital agencies. Triggers on requests like "shortlist three B2B branding agencies in California", "find a PPC shop with ecommerce experience", "we need a content marketing partner for a SaaS launch", or "pull contact info for these 12 agency domains", even when the need is described indirectly. Drives the ServiceGraph API (api.servicegraph.co) — a 100k+ US firm catalog filterable by industry, services, location, size, ratings, and third-party listings. Skip SEO-only asks (use find-seo-agency), web/software-development asks (use find-web-developer or find-software-developer), recruiting an in-house marketing hire, "write me a marketing plan" do-the-work asks, non-US firms, individual freelancers, marketing-software-product recommendations, and consumer/personal-brand asks.
Use whenever the user wants to find, shortlist, vet, or enrich US marketing agencies — including branding, content marketing, PPC/paid media, social media…
As a process C 59/100 · Has gaps — 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 instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
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 · 1
✓ No critical or high findings
Medium and low: 1
-
medium Exfiltration
net-credential-useSKILL.md:85Credential used in a network call (verify the destination is the intended service)curl -sS -H "Authorization: Bearer $SERVICEGRAPH_API_KEY" \
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 59/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 8 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 20 steps
- 100Failures and branches. 2 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3068 tokens
- low 10 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1015: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 5 example trigger phrases
- +4Structure: 19 headings
- +3Step-by-step instructions: 20 items
- +4Has examples (12 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.