Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This skill library guides direct response campaigns from the buyer's psychology up: extracting avatars and offers, then building headlines, mechanisms, ad angles, and conversion paths from that foundation. It helps copywriters, paid media buyers, and marketing strategists work from a shared strategic core so the creative doesn't get ahead of the positioning.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
I’m using
Install the Advertising Skills bundle manually
Installs: Complete Advertising Skills library · Claude Code, Cursor, Windsurf, Codex, custom agent frameworks, Open Claw
Before you start
An agent that supports the Agent Skills pattern
Manually place the repository's skills in the agent skills directory `.agents/skills/`, preserving the documented `skills/<category>/<skill-name>/SKILL.md` folder structure.
Any available customer language, research, or observations
Use the avatar-extraction skill to define one specific primary buyer for my offer. Based only on the buyer and offer information I provide, describe their current situation, pain points, desired outcome, failed attempts, and emotional drivers. Use real-world language, focus on their decision context, and flag any important information that is missing instead of guessing.
Use the generic-language-killer skill to review the short marketing copy I provide. Identify vague phrases, fluff, and buzzwords, then rewrite the copy with specific, direct, human wording while preserving its factual meaning and avoiding a corporate tone.
Use the performance-diagnosis skill to diagnose this underperforming campaign. Check the offer, audience, creative, funnel, and sales information I provide, then identify the best-supported root cause and recommend one focused fix. Clearly state where the available evidence is insufficient rather than guessing.