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 code-yeongyu/oh-my-openagent --skill ultimate-browsinggit clone --depth 1 https://github.com/code-yeongyu/oh-my-openagentWrote 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/code-yeongyu/oh-my-openagent/ultimate-browsing)<a href="https://agentmods.dev/skills/code-yeongyu/oh-my-openagent/ultimate-browsing"><img src="https://agentmods.dev/badge/skills/code-yeongyu/oh-my-openagent/ultimate-browsing/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/code-yeongyu/oh-my-openagent/ultimate-browsing"><img src="https://agentmods.dev/badge/skills/code-yeongyu/oh-my-openagent/ultimate-browsing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high YARA Match · line 29 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
- medium Rogue Agent · line 123 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00071 | $0.02716 |
| Opus 5.5 | $0.00028 | $0.01086 |
| Sonnet 5.5 | $0.00014 | $0.00543 |
| Haiku 4.5 | $0.00007 | $0.00272 |
Grade C, and why
ultimate-browsing scanned grade C with 2 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 11d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
await browser.close() // then rm -rf the profile Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s "https://r.jina.ai/https://weibo.com/<uid>/<pid>" # weibo via Jina The source is not reproduced here
A licence we could not identify
The repository carries a LICENSE file, but it is custom or dual enough that GitHub cannot name it and neither can this catalogue. Unknown terms are not permission, so the body is not copied here. Read the licence at the source and decide for yourself.
What ships with it
57 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.
- .gitignore 29 B
- ATTRIBUTION.md 2.5 KB
- engine/__init__.py 668 B runs code
- engine/__main__.py 4.8 KB runs code
- engine/AGENTS.md 10 KB
- engine/bias_check.py 7.1 KB runs code
- engine/curl_probe.py 2.4 KB runs code
- engine/executor.py 6.3 KB runs code
- engine/fetch_chain.py 13 KB runs code
- engine/referers.py 327 B runs code
- engine/result_schema.py 1.5 KB runs code
- engine/summary.py 1.2 KB runs code
- engine/surrogate.py 7.9 KB runs code
- engine/surrogates.yaml 2.3 KB
- engine/templates/package.json 349 B
- engine/templates/playwright_mobile_chrome.js 3.6 KB runs code
- engine/templates/playwright_real_chrome.js 5.6 KB runs code
- engine/tests/fixtures/amp_redirect_stub.html 323 B
- engine/tests/fixtures/search_interstitial.html 90 KB
- engine/tests/fixtures/wayback_available.json 246 B
- engine/tests/fixtures/wayback_snapshot.html 181 KB
- engine/tests/test_fetch_chain.py 2.7 KB runs code
- engine/tests/test_playwright_stealth.py 3.9 KB runs code
- engine/tests/test_playwright_templates.py 11 KB runs code
- engine/tests/test_surrogate_validators.py 3.5 KB runs code
- engine/tests/test_surrogate.py 11 KB runs code
- engine/url_transforms.py 3.2 KB runs code
- engine/validators.py 9.1 KB runs code
- engine/waf_detector.py 6.9 KB runs code
- engine/waf_profiles.yaml 6.2 KB
- references/agent-reach/career.md 712 B
- references/agent-reach/dev.md 1.3 KB
- references/agent-reach/README.md 2.2 KB
- references/agent-reach/search.md 821 B
- references/agent-reach/social.md 5.6 KB
- references/agent-reach/video.md 2.7 KB
- references/agent-reach/web.md 1.9 KB
- references/chrome-stealth.md 2.2 KB
- references/insane-search/cache-archive.md 4.1 KB
- references/insane-search/fallback.md 6.0 KB
- references/insane-search/jina.md 3.8 KB
- references/insane-search/json-api.md 3.5 KB
- references/insane-search/media.md 3.5 KB
- references/insane-search/metadata.md 2.9 KB
- references/insane-search/naver.md 3.3 KB
- references/insane-search/playwright.md 6.7 KB
- references/insane-search/public-api.md 3.4 KB
- references/insane-search/README.md 20 KB
- references/insane-search/rss.md 2.9 KB
- references/insane-search/tls-impersonate.md 6.3 KB
- references/insane-search/twitter.md 3.5 KB
- scripts/cookie_crypto.py 2.9 KB runs code
- scripts/cookie_domains.py 1.1 KB runs code
- scripts/cookie_paths.py 3.8 KB runs code
- scripts/extract_cookies.py 9.3 KB runs code
- scripts/tests/test_cookie_domain_filter.py 3.5 KB runs code
- scripts/tests/test_extract_cookies.py 9.4 KB runs code
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.
- 11d ago Changed · -2 lines · +6 tokens per session ff807c0ae8f3
- 29d ago Changed · +7 lines scan A → C c42b2faeb084
- 1mo ago Changed · -12 lines 535ea8a233c7
- 1mo ago Changed · +12 tokens per session ff224d16a9d0
- 1mo ago Changed · +15 lines · -207 tokens per session 49598aeae816
- 1mo ago First seen · 143 lines · 260 tokens per session scan A 071f3134254f
ultimate-browsing is a skill published in the GitHub repository code-yeongyu/oh-my-openagent (70,278 stars, last pushed today), with no licence file. It adds 71 tokens to every session and 2,716 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
e2e-testing-patterns
Master end-to-end testing with Playwright and Cypress to build reliable test suites that catch bugs, improve confidence, and enable fast deployment. Use when implementing E2E tests, debugging flaky tests, or establishing testing standards.
ai-web-scraping-scrapegraph
AI-powered web scraping - extract data using natural language prompts.
browser-automation-notte
Browser automation - control browser sessions, scrape pages, and run AI agents.
extract-webpage-data
Extract structured data from web pages using AI.
browser-test-loop
Use to actually verify a web change in a real browser instead of assuming it works — click through a flow, catch console errors, check a form submits, confirm a fix on the page. Trigger when a UI change needs proof, a bug only reproduces in the browser, or the user asks to test, check, or screenshot the running app.
qa
Exercise the delivered customer journey using real runtime or browser evidence. Use for functional acceptance testing after implementation, including failure paths.