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 KunanonJ/ai-skills-hub --skill aside-google-searchgit clone --depth 1 https://github.com/KunanonJ/ai-skills-hubWrote 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/kunanonj/ai-skills-hub/aside-google-search)<a href="https://agentmods.dev/skills/kunanonj/ai-skills-hub/aside-google-search"><img src="https://agentmods.dev/badge/skills/kunanonj/ai-skills-hub/aside-google-search/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/kunanonj/ai-skills-hub/aside-google-search"><img src="https://agentmods.dev/badge/skills/kunanonj/ai-skills-hub/aside-google-search.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 | $0.00021 | $0.00639 |
| Opus 5 | $0.00010 | $0.00319 |
| Sonnet 5 | $0.00004 | $0.00128 |
| Haiku 4.5 | $0.00002 | $0.00064 |
Grade A, and why
aside-google-search 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 9d 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 — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Google Search
Use the googleSearch global in the REPL tool. It fetches Google Search with the browser profile's cookies, then parses the result DOM with stable selectors verified against a live SERP. You do not need to manually inspect the results page unless parsing fails.
Quick Reference
const results = await googleSearch.search('openai', { limit: 5 });
console.log(JSON.stringify(results, null, 2));
// → [{ title, url, sourceName, publishedAtText, snippet, sitelinks }]
// Pagination
const page2 = await googleSearch.search('openai', { start: 10 });
const page3 = await googleSearch.search('openai', { start: 20 });
Query Discipline
Do not run Google Search queries in parallel. Parallel query bursts can trigger CAPTCHA or bot blocks even when each individual query is valid.
Avoid patterns like:
const queries = ['first query', 'second query', 'third query'];
await Promise.all(queries.map((q) => googleSearch.search(q)));
Also avoid firing multiple searches from loop bodies without waiting for each result before deciding the next query. Keep Google searches sequential and only issue the next query after you have inspected the previous result.
Methods
googleSearch.search(query: string, opts?: GoogleSearchOptions): Promise<GoogleSearchResult[]>
Search Google web results and return compact structured results.
- The parser prefers stable attrs observed in the live DOM:
- result blocks:
div[data-rpos] - primary title/link: first external
a[href]that wraps anh3
- result blocks:
- This intentionally avoids relying on Google's churn-heavy class names and internal
jsnamehooks. - Pagination uses Google's
startoffset:- page 1 = omit
startor usestart: 0 - page 2 =
start: 10 - page 3 =
start: 20
- page 1 = omit
- If the helper fails, open the Google Search URL directly and inspect the page.
Types
interface GoogleSearchOptions {
limit?: number; // max results. default: 10
safeSearch?: 'active' | 'off';
language?: string; // e.g. 'en', 'ko', 'ja'
country?: string; // e.g. 'us', 'kr', 'jp'
start?: number; // pagination offset. e.g. 10 = page 2
time?: 'day' | 'week' | 'month' | 'year';
}
interface GoogleSearchResult {
title: string;
url: string;
sourceName?: string;
publishedAtText?: string; // e.g. '2 hours ago'
snippet?: string;
sitelinks?: Array<{ title: string; url: string }>;
}
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.
- 9d ago First seen · 74 lines · 21 tokens per session scan A 11bf23344f9f
aside-google-search is a skill published in the GitHub repository KunanonJ/ai-skills-hub (5 stars, last pushed 2d ago), licensed MIT. It adds 21 tokens to every session and 639 once invoked, about $0.0001 per session on Opus 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 2026-09-03.
Other skills, from other repositories
issue-creation
Trigger: issue creation, bug reports, feature requests, or issue approval. Create and triage GitHub issues from repository evidence.
sdd-tasks
Break an SDD change into implementation tasks. Trigger: orchestrator launches task planning for a change.
work-unit-commits
Plan commits as reviewable work units. Trigger: implementation, commit splitting, chained PRs, or keeping tests and docs with code.
systemic-issue-triage
Trigger: new issue, bug report, triage, backlog, issue flood, community report, root cause, dead-end, blocked user. Attack issues by root class, never one-by-one; fixes must shrink the system, not grow it.
sdd-research
Trigger: SDD research, external evidence, source-backed research. Produce auditable evidence for a selected research lane.
skill-improver
Trigger: improve skills, audit skills, refactor skills, skill quality. Audit and upgrade existing LLM-first skills.