Checks every claim in a piece of copy (ads, emails, posts, pages, scripts, briefs) against the claims the brand has approved. Keeps, rewrites or cuts each line, and lists the proof that would bring a cut line back. When a line the copy needs has no approval, it asks for a short sign-off on the 3 or 4 lines that matter and records the answer. Use alone when you already have the copy and a brand set up; otherwise run the playbook for the job (e.g. make-static-ads, make-content, write-outreach).
84%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow reads brand-owned learnings, the Brand record, and the human's chat answers, but it also checks copy spans that callers extract from pages, which can include arbitrary web content, so under recall-first it is scored low rather than no.
gooseworks-ai/goose-skills
885edb5
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.