adspirer-skills

v0.1.0

Published

2 months ago

Install Adspirer Skills — expert Google, Meta, TikTok & LinkedIn ad-ops workflows — into any AI agent (Claude Code, ChatGPT, Cursor, Codex, Gemini, and more). Open Agent-Skills (SKILL.md) format.

adspirer-skills

Install Adspirer Skills — 92 expert advertising workflows for AI agents — with one command.

Ready-to-run skills for Google Ads, Meta (Facebook/Instagram) Ads, TikTok Ads, LinkedIn Ads, and Amazon Ads: media planning, creative production, campaign launch, and always-on ad ops. Each is a bounded workflow in the open Agent Skills (SKILL.md) format — so it drops into Claude Code, Claude, ChatGPT, Cursor, Codex, Gemini CLI, and any Agent-Skills-compatible AI.

npx adspirer-skills list                                     # browse all 92 skills
npx adspirer-skills add the-daily-pacing-check                   # install (Claude Code by default)
npx adspirer-skills add meta-full-funnel-launcher --agent cursor
npx adspirer-skills search "meta campaign"

No account needed to install or run. Browse everything at adspirer.com/skills.

Agents

claude-code (default) · claude · cursor · codex · windsurf · gemini · openclaw · antigravity · hermes · other

npx adspirer-skills add <skill> --agent <name>

The CLI writes each skill into your agent's native location (Claude Code → .claude/skills/, Cursor → .cursor/rules/, Codex → AGENTS.md, and so on).

Why these are safe to run

Every Adspirer Skill carries the same contract: fixed checks against live account data, honest stop conditions, and approval gates on anything that spends. Skill content is reference material — authority comes only from you, and nothing spends a dollar without your explicit approval. The ads leg connects through Adspirer only when a step touches a live ad account.

Popular skills

A/B Test Runner — Set up a clean A/B test (one variable, proper split, a decision rule and duration for significance), read it honestly, and call the winner only when the data supports it.
Ad Copy Writing Room — A copy-writing prompt that produces platform-correct ad copy variations (headlines, primary text, CTAs) grounded in a real offer and audience, with angle rationale.
Ad-Library Deep Dive — Systematically mine a brand's Meta Ad Library: creative angles, hooks, formats, longevity (long-running = winning) — extract patterns to inform your own creative.
Agency New-Client Onboarding — Onboard a new client end to end: audit their accounts, set up tracking, build the initial plan, launch the first campaigns, and establish the reporting cadence.
Amazon Sponsored Products Launcher — Launch an Amazon Sponsored Products campaign: product eligibility, keyword/ASIN targeting, bids, negatives, built and checked.
Anomaly Triage — When a metric moves sharply, triage it: is it real, a tracking artifact, a platform delay, or a genuine change? Trace it to a driver before anyone panics.
App Campaign (UAC) Launcher — Launch a Google App campaign (UAC): conversion events/SDK check, asset variety, bidding by install vs in-app action, built paused and verified.
App Growth Stack — Stand up an app-growth stack: UAC + Meta app campaigns, install + in-app event tracking, cohort/LTV read, and a scaling plan — end to end, gated.
Audience Architect — Design a layered audience strategy (cold/warm/hot, exclusions, lookalikes/custom) for a campaign, per platform, with overlap and audience-size sanity checks.
Avatar Spokesperson Ads — Generate avatar-spokesperson video ads via HeyGen — script, avatar, voiceover — at scale for testing message variations.
B2B Media Plan (LinkedIn-first) — Build a B2B media plan centered on LinkedIn: firmographic targeting, ABM lists, lead-gen vs traffic objectives, realistic B2B CPLs, and budget staging.
B2B Pipeline Engine — Build a B2B pipeline engine: LinkedIn + Search lead gen → CRM sync (HubSpot if connected) → follow-up → attribute pipeline back to campaigns.