ads-budget — Claude Code skill: what it does, how to install
Plans and audits paid-media spend, bidding strategies, and pacing across channels. Useful when balancing marginal returns and cash constraints.
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a claude-codeWhat ads-budget does
The skill structures paid acquisition planning into a step-by-step workflow. It starts by gathering conversion value, gross margin, cash ceilings, and attribution uncertainty. It normalizes spend windows across platforms before comparing performance.
Instead of relying on blended averages, the skill looks at marginal evidence. The resulting plan specifies exact amounts, owners, guardrails, success metrics, and explicit rollback triggers for each campaign. Rules like 70/20/10 or fixed CPA multiples are treated only as optional heuristics.
The workflow relies on ads-math for calculations and forecasts. Without baseline performance metrics and margin data, the skill will not authorize budget changes.
What it helps with
- calculate break-even boundaries with sensitivity cases and formulas
- separate committed baseline spend from experiments and reserve capacity
- evaluate scaling and reallocation options using marginal return evidence
- establish ad spend pacing plans with rollback triggers and guardrails
Example prompts for the agent
- Review our ad spend pacing and recommend allocation changes based on marginal ROAS.
- Create a media budget plan with cash constraints, break-even thresholds, and rollback triggers.
- Compare scaling versus reallocating spend across platforms using our gross margin data.
How to install ads-budget
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a claude-code2Manually: put the files in place
~/.claude/skills/ads-budget/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/ads-budget/
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a codex2Manually: put the files in place
.agents/skills/ads-budget/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a cursor2Manually: put the files in place
.cursor/skills/ads-budget/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install AgriciDaniel/claude-ads ads-budget2Manually: put the files in place
.github/skills/ads-budget/SKILL.md1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a gemini-cli2Platform CLI
gemini skills install https://github.com/AgriciDaniel/claude-ads --path skills/ads-budgetManually: put the files in place
.devin/skills/ads-budget/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-ads --skill ads-budget -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
From the same set20
The whole set →Similar
Questions about ads-budget
How to install ads-budget in Claude Code?
Run npx skills add AgriciDaniel/claude-ads --skill ads-budget -a claude-code in your terminal.
What does ads-budget do?
It guides ad spend planning and review by testing break-even boundaries, marginal returns, and risk guardrails across campaigns.
Are additional tools required for forecasting?
Yes, the skill instructions specifically point to ads-math for handling forecasts and calculations.
Source text
SKILL.md · MIT
--- name: ads-budget description: "Plan and review paid-media budgets, bidding, pacing, marginal return, targets, constraints, and allocation across supported platforms. Use for ad budget allocation, media budget, bidding strategy, scaling, spend pacing, ROAS target, or investment tradeoffs. Use ads-math for calculations and forecasts." --- # Budget and Bidding 1. Establish objective, conversion value, gross margin, cash constraints, sales capacity, attribution uncertainty, seasonality, and platform minimum evidence. 2. Normalize spend and outcomes to comparable windows and definitions. 3. Calculate break-even boundaries and show formulas, inputs, uncertainty, and sensitivity cases. 4. Distinguish committed baseline, controlled experiments, and reserve capacity. 5. Compare hold, reallocate, scale, reduce, or experiment options using marginal evidence rather than blended averages alone. 6. Return a decision-complete plan with platform/campaign amount, timing, owner, guardrails, success measure, and rollback trigger. Rules such as 70/20/10, fixed CPA multiples, fixed budget-to-CPA ratios, and fixed percentage scaling are optional heuristics, never universal authorization.
© AgriciDaniel · MIT · AgriciDaniel/claude-ads