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 sergeyizmailov/knowledge-delta-skills --skill google-adsgit clone --depth 1 https://github.com/sergeyizmailov/knowledge-delta-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/sergeyizmailov/knowledge-delta-skills/google-ads)<a href="https://agentmods.dev/skills/sergeyizmailov/knowledge-delta-skills/google-ads"><img src="https://agentmods.dev/badge/skills/sergeyizmailov/knowledge-delta-skills/google-ads/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/sergeyizmailov/knowledge-delta-skills/google-ads"><img src="https://agentmods.dev/badge/skills/sergeyizmailov/knowledge-delta-skills/google-ads.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 | $0.00151 | $0.02198 |
| Opus 5.5 | $0.00060 | $0.00879 |
| Sonnet 5.5 | $0.00030 | $0.00440 |
| Haiku 4.5 | $0.00015 | $0.00220 |
Grade A, and why
google-ads 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Google Ads
Operate as a senior Google Ads practitioner. Treat UI labels, eligibility, policy, API behavior, prices, and benchmarks as volatile. Research reviewed 2026-08-27; hub dates re-verified 2026-09-03. Baseline 2026-09-03 (Sonnet 5 blind, 4 tasks): 3 [W] (AI Max opt-out, Search-vs-PMax rule, MC link services), 1 [R] (API mutate basics); rerun with skills fixed 6/6 flagged items. Verify primary sources whenever the answer depends on current behavior.
Cross-platform note: this pairs with meta-ads/meta-grey-ops. Do not port Meta structure doctrine
to Google. Google is explicit-search-intent driven — Menachem Ani: "every search query is a person
telling you something, not a demographic."
The rule is not "don't consolidate" — it is "never merge different intent or different economics."
Consolidation is how you clear Smart Bidding's volume floor (01); Hagakure-style consolidated
structures are standard on Google. What fails is Meta-style consolidation across price tiers,
margins, funnel stages, because Google can read the intent Meta cannot. Ani prescribes segmentation by
economic/price tier, isolated retargeting, product-theme grouping — not fewer campaigns. See 05.
Check these two things first on any 2026 account
- The AI Max forced migration (
05). Campaign-level broad match and Automatically Created Assets auto-convert 2026-09-01 → 09-30, no rollback; migration is in progress as of this review (confirmed 2026-09-03). The opt-out window (disabling the legacy setting before Sep 1) has closed. - The 2026-08-17 budget-limited target enforcement (
02). Any tCPA/tROAS drift on a "Limited by budget" campaign since that date is expected behavior, not a tracking bug [confirmed 2026-09-03].
Route references
Read only what the task needs.
| Need | Reference |
|---|---|
| Evidence rules, refuted claims, volatile list, intake | references/00-evidence-and-maintenance.md |
| Campaign types, Search vs PMax vs Shopping vs Demand Gen vs App picker, MCC, arbitration | references/01-account-architecture.md |
| Bid strategies, Ad Rank, Quality Score, budgets, Auction Insights | references/02-bidding-auction-quality.md |
| Match types, prioritization, search terms, negatives, n-grams, competitive intel | references/03-keywords-and-negatives.md |
| RSA specs, Ad Strength, pinning, assets, ad testing, dynamic text, auto-assets | references/04-creative-and-assets.md |
| AI Max migration, AI Overviews, what survives September | references/05-ai-max-and-ai-surfaces.md |
| Conversion actions, click IDs, enhanced conversions, OCI, consent, sGTM, attribution | references/06-tracking-attribution.md |
| PMax, Demand Gen, audiences, cannibalization, failure modes | references/07-pmax-demand-gen-audiences.md |
| Benchmarks, unit economics, impression-share math, diagnostic tree, columns | references/08-benchmarks-diagnostics.md |
| Policy taxonomy, enforcement tracks, verification, certifications, destination rules | references/09-policy-and-compliance.md |
| API access, GAQL, mutates, mass launch, Scripts, AI pipelines | references/10-api-and-automation.md |
| API errors: code → cause → fix | references/11-api-error-catalog.md |
| Operator edge — ratcheting, signal engineering, wasted-spend hunts, dead techniques | references/12-operator-edge.md |
What ships with it
16 files 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.
- playbooks/b2b-saas.md 4.6 KB
- playbooks/local-leadgen.md 4.9 KB
- playbooks/us-ecommerce.md 5.1 KB
- references/00-evidence-and-maintenance.md 5.5 KB
- references/01-account-architecture.md 15 KB
- references/02-bidding-auction-quality.md 16 KB
- references/03-keywords-and-negatives.md 15 KB
- references/04-creative-and-assets.md 14 KB
- references/05-ai-max-and-ai-surfaces.md 12 KB
- references/06-tracking-attribution.md 18 KB
- references/07-pmax-demand-gen-audiences.md 19 KB
- references/08-benchmarks-diagnostics.md 18 KB
- references/09-policy-and-compliance.md 20 KB
- references/10-api-and-automation.md 23 KB
- references/11-api-error-catalog.md 8.5 KB
- references/12-operator-edge.md 20 KB
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.
- 1mo ago Changed · +6 lines · +44 tokens per session 15fca6e7a69f
- 1mo ago First seen · 143 lines · 107 tokens per session scan A c7ad07c500a0
google-ads is a skill published in the GitHub repository sergeyizmailov/knowledge-delta-skills (3 stars, last pushed 1mo ago), licensed MIT. It adds 151 tokens to every session and 2,198 once invoked, about $0.0006 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.
Other skills, from other repositories
mem0-status
Diagnoses mem0 connectivity, API key validity, and memory read/write functionality. Use when memory operations fail, searches return empty, addmemory errors occur, or to verify the plugin is working correctly.
error-handling-patterns
Master error handling patterns across languages including exceptions, Result types, error propagation, and graceful degradation to build resilient applications. Use when implementing error handling, designing APIs, or improving application reliability.
diagnose-backend-bug
Diagnose a bounded backend or multi-service failure from GitHub Issues, Jira, Aone, user-provided exports, logs, traces, responses, stack traces, or job records. Use when a service, API, RPC, worker, queue, CLI, or scheduled job bug needs correlation through the project's existing observability route before repair; do…
firebase-auth
Use when setting up auth, managing auth state, implementing email/password or social sign-in, handling auth errors, or managing users.
firebase-crashlytics
Use when implementing crash reporting, capturing fatal/non-fatal errors, recording isolate/async exceptions, customizing reports, or uploading obfuscated symbols.
developing-genkit-dart
Use when building AI agents in Dart, implementing Genkit flows or tools, integrating LLMs into Dart or Flutter applications, or using Genkit Dart plugins.