Hainrixz/claude-ads / ads
ads
UnverifiedFree-first paid advertising audit and optimization skill, focused on the 3 platforms where 95% of advertiser spend lives: Meta, Google, TikTok. Guided onboarding (/ads start) saves context once and step-by-step OAuth walkthroughs; continuous coach (/ads next) ranks Quick Wins after each audit. ~158 weighted checks with scoring, parallel agents, industry templates, and AI creative generation.
references/update-cost-warning.md
View on GitHub# `/ads update` — Cost & Credit Warning
> **Read this before running `/ads update`. It is the most credit-intensive command in claude-ads.**
`/ads update` refreshes per-platform reference files with the last 30 days of changes by aggregating sources from:
- **Reddit JSON API** (free, no key) — practitioner subreddits per platform
- **Hacker News Algolia API** (free, no key) — keyword-filtered to ad-tech topics
- **Official platform changelogs** — vendor release notes (Google Ads, Meta Marketing API, TikTok / LinkedIn / Microsoft / Apple Ads)
- **Web search** — industry press fallback (Search Engine Land, AdWeek, etc.)
For each platform, this means **20–50 outbound HTTP requests** plus LLM synthesis to produce the digest.
## Estimated cost per run
| Mode | Platforms covered | Approx. tokens | Approx. minutes |
|---|---|---|---|
| `/ads update <one platform>` | 1 | ~50,000–150,000 | 2–5 min |
| `/ads update all` | 7 (loops sequentially) | ~500,000+ | 15–30 min |
Estimates assume `--depth default`. Quick mode (`--depth quick`) is roughly **40% of the above**; deep mode (`--depth deep`) is roughly **2× the above**.
## Recommendations
1. **If you're on a low-credit plan**, run per platform — never `all`.
2. **Use Sonnet, not Opus**, for `/ads update` runs. Opus is overkill for synthesis-from-citations work and costs ~5× more.
3. **Refresh monthly, not daily.** The reference data is dated `Last 30 Days`; rerunning the next day produces near-identical output and burns credits for nothing.
4. **Cancel is always safe.** The skill writes per-platform files atomically — interrupting mid-`all` leaves earlier platforms' digests intact.
## Confirmation
The skill will ask for explicit confirmation before fetching. You can:
- **Proceed** — run the requested mode at the cost estimated above
- **Switch to quick depth** — same coverage, ~40% of the cost
- **Cancel** — abort with no fetches made
Pick **Cancel** if you're unsure. References are still fully usable without an `/ads update` run.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |