Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kelpi-ai/meta-ads-skills --skill audience-buildergit clone --depth 1 https://github.com/kelpi-ai/meta-ads-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kelpi-ai/meta-ads-skills/audience-builder)<a href="https://agentmods.dev/skills/kelpi-ai/meta-ads-skills/audience-builder"><img src="https://agentmods.dev/badge/skills/kelpi-ai/meta-ads-skills/audience-builder/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kelpi-ai/meta-ads-skills/audience-builder"><img src="https://agentmods.dev/badge/skills/kelpi-ai/meta-ads-skills/audience-builder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00043 | $0.00442 |
| Opus 5.5 | $0.00017 | $0.00177 |
| Sonnet 5.5 | $0.00009 | $0.00088 |
| Haiku 4.5 | $0.00004 | $0.00044 |
Grade A, and why
audience-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Audience Builder
Doctrine
Targeting is mostly the algorithm's job now: Meta removed exclusion options in 2025, retired interest categories in January 2026, and Andromeda aims by reading the ad. Audience work in 2026 is (a) first-party assets (customer lists, lookalikes) that give the system a seed, and (b) knowing when broad is simply better. An honest audience skill often ends with "go broad and fix the angle instead."
When to use
- Before launch, if you have a customer list worth seeding.
- When delivery lands in the wrong crowd ("my ad for young women went to older men"): usually an angle problem wearing a targeting costume.
- Needs a Meta Ads MCP (read for research, write only for creating audiences you approve).
Run it
Audience research for my offer: [ONE-LINE OFFER + WHO from my angles]
1. Search interests and behaviors related to my category. For each, give the estimated audience size.
2. If I have a customer list or purchasers: propose a custom audience plus a 1% lookalike, with estimated sizes. Draft only, do not create yet.
3. Then be honest with me: given my budget of [$X/day], would broad targeting likely beat these? Explain in two sentences using what my ad creative already signals about who it is for.
4. Wait for my pick before creating anything.
Guardrails
- Research is read-only; audience creation happens only on explicit approval.
- Never stack interests to feel in control. If the recommendation is broad, say broad.
- If an ad set was built on since-retired targeting options, flag it: those stopped delivering after January 15, 2026.
Good output looks like
A short table of real options with sizes, one honest recommendation, and no audience created that you did not ask for by name.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 34 lines · 43 tokens per session scan A 0a21fe20a668
audience-builder is a skill published in the GitHub repository kelpi-ai/meta-ads-skills (5 stars, last pushed 4mo ago), licensed MIT. It adds 43 tokens to every session and 442 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
audience-builder compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| audience-builderkelpi-ai/meta-ads-skills | Skill | 5 | 4mo ago | MIT | A | 43 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 314k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |