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 askads/mcp-vk-ads --skill vk-ads-auditgit clone --depth 1 https://github.com/askads/mcp-vk-adsWrote 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/askads/mcp-vk-ads/vk-ads-audit)<a href="https://agentmods.dev/skills/askads/mcp-vk-ads/vk-ads-audit"><img src="https://agentmods.dev/badge/skills/askads/mcp-vk-ads/vk-ads-audit/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/askads/mcp-vk-ads/vk-ads-audit"><img src="https://agentmods.dev/badge/skills/askads/mcp-vk-ads/vk-ads-audit.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 | $0.00000 | $0.00277 |
| Opus 5.5 | $0.00000 | $0.00111 |
| Sonnet 5.5 | $0.00000 | $0.00055 |
| Haiku 4.5 | $0.00000 | $0.00028 |
Grade A, and why
vk-ads-audit 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 6d 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
VK Ads MCP
What this server covers
VK Ads campaigns, ad groups, ads and their statistics. Reading is safe; budget and bid changes spend real money.
Before the first call
Confirm the token works and the account is the one you expect. A token scoped to an agency account reports the agency's own objects, not the client's.
Pulling statistics
Request one wide period instead of looping day by day or campaign by campaign. Each call costs quota, and a single wide request returns the same numbers as many narrow ones.
Money values come back in the account currency. Do not mix currencies when you sum across accounts.
Paging
Prefer a large limit over many small pages. When a list comes back truncated, the server
says so explicitly rather than hiding it — carry on paging instead of treating the partial
list as complete.
Writes
Budget and bid changes take effect immediately and are not reversible by the server. State the current value, the new value and the affected object before applying a change, and apply changes one object at a time so a partial failure is readable.
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.
- 6d ago First seen · 38 lines · 0 tokens per session scan A 39e3f6c3032c
vk-ads-audit is a skill published in the GitHub repository askads/mcp-vk-ads (9 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 277 tokens. 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 2026-10-04.
Other skills, from other repositories
budget-optimizer
Optimize budget allocation across ad platforms and campaigns. Use when the user asks about budget distribution, media planning, funnel-stage budgeting, or wants spend reallocation recommendations.
campaign-manager
Manage ad campaigns across platforms. Use when the user wants to create, modify, pause, enable, or check the status of advertising campaigns on Google Ads, Meta, LinkedIn, Reddit, Microsoft Ads, or YouTube.
competitor-research
Research competitor advertising strategies, keywords, ad copy, and landing pages. Use when the user asks about competitor PPC spend, keyword gaps, or wants to analyze competing brands.
creative-generator
Generate ad creatives including display images, video ads, and copy variations. Use when the user needs ad images, headlines, descriptions, or wants to A/B test creative variations.
performance-analyzer
Analyze campaign performance metrics, compare ads, and track attribution. Use when the user asks about campaign results, ROAS, CPA, conversion rates, or wants performance reports.
pkf-writing
Author, update, convert, validate, persist, and render PKF (.pkf) content through message-inline, Library Asset, or Memory Page carriers. Use for structured reports, knowledge pages, delivery cards, or any .pkf body. NOT for recall, memory-graph placement, or curation decisions — those belong to the memory skill.