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 ogulcancelik/pi-extensions --skill pi-web-browsegit clone --depth 1 https://github.com/ogulcancelik/pi-extensionsWrote 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/ogulcancelik/pi-extensions/pi-web-browse)<a href="https://agentmods.dev/skills/ogulcancelik/pi-extensions/pi-web-browse"><img src="https://agentmods.dev/badge/skills/ogulcancelik/pi-extensions/pi-web-browse/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/ogulcancelik/pi-extensions/pi-web-browse"><img src="https://agentmods.dev/badge/skills/ogulcancelik/pi-extensions/pi-web-browse.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
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 →
- medium MCP Rug Pull · line 23 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00047 | $0.00877 |
| Opus 5.5 | $0.00019 | $0.00351 |
| Sonnet 5.5 | $0.00009 | $0.00175 |
| Haiku 4.5 | $0.00005 | $0.00088 |
Grade A, and why
pi-web-browse scanned grade A with 1 finding 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 2mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
description: "Search the web and fetch/read pages via a real headless browser (CDP). Use this instead of curl when sites are JS-heavy or bot-protected. Works on Linux, macOS, and Windows." How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Web Browse
Search the web, then open/fetch pages in a real browser session (headless Chromium via CDP) and extract readable text.
Use this instead of curl when sites are JS-heavy or bot-protected.
Setup
Run once before first use:
cd {baseDir}
npm install
The skill auto-detects browsers already installed on your system (Brave, Chrome, Edge, Chromium). On Windows, Edge is pre-installed and works out of the box.
No browser installed? (rare) Run: npx playwright install chromium
Configuration (optional)
Environment variables:
| Variable | Description |
|---|---|
WEB_BROWSE_BROWSER_BIN |
Path to browser binary (auto-detected if not set) |
WEB_BROWSE_USER_AGENT |
Override User-Agent string |
WEB_BROWSE_DAEMON_PORT |
Daemon port (default: 9377) |
WEB_BROWSE_CDP_PORT |
CDP port (default: 9225) |
WEB_BROWSE_DEBUG_DUMP |
Set to 1 to save screenshots/HTML on failures |
You can also pass --browser-bin <path> as a CLI argument.
Usage
# Search (results are cached for ~10 minutes)
{baseDir}/web-browse.js "your query"
{baseDir}/web-browse.js "your query" -n 10
# Fetch specific cached results by index
{baseDir}/web-browse.js --fetch 1,3,5
# Fetch a specific URL
{baseDir}/web-browse.js --url <url> # truncated (~2000 chars)
{baseDir}/web-browse.js --url <url> --full # full content
Windows note: Use node {baseDir}/web-browse.js instead of {baseDir}/web-browse.js
Default behavior: persistent daemon (auto)
Direct calls automatically start/use a local daemon that keeps a persistent headless browser+CDP session. This avoids browser startup overhead and helps with bot-protection pages that auto-clear (e.g. Anubis PoW).
Daemon controls (optional, for debugging)
{baseDir}/web-browse.js --daemon status
{baseDir}/web-browse.js --daemon start
{baseDir}/web-browse.js --daemon stop
{baseDir}/web-browse.js --daemon restart
Bypass daemon (one-shot)
What ships with it
22 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 170 B
- lib/bot-protection.js 2.9 KB runs code
- lib/browser-bin.js 4.7 KB runs code
- lib/cdp.js 6.0 KB runs code
- lib/daemon-client.js 4.9 KB runs code
- lib/daemon.js 15 KB runs code
- lib/debug-dump.js 1.5 KB runs code
- lib/extract.js 1.8 KB runs code
- lib/fetch.js 1.9 KB runs code
- lib/fingerprint.js 11 KB runs code
- lib/http-fetch.js 1.2 KB runs code
- lib/search.js 11 KB runs code
- LICENSE 1.0 KB
- package.json 1.4 KB
- README.md 2.8 KB
- test/bot-protection.test.js 609 B runs code
- test/browser-bin.test.js 1.4 KB runs code
- test/daemon-client.test.js 1.9 KB runs code
- test/extract.test.js 1.3 KB runs code
- test/fingerprint.test.js 565 B runs code
- test/search-ddg.test.js 766 B runs code
- web-browse.js 22 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.
- 2mo ago First seen · 102 lines · 47 tokens per session scan A 233bce123094
pi-web-browse is a skill published in the GitHub repository ogulcancelik/pi-extensions (570 stars, last pushed 2d ago), licensed MIT. It adds 47 tokens to every session and 877 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
pi-web-browse compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| pi-web-browseogulcancelik/pi-extensions | Skill | 570 | 2d ago | MIT | A | 47 | Claude Code, Codex |
| interactive-loginyc-software/qm | Skill | 16k | today | MIT | A | 46 | Claude Code, Codex |
| pinchtab-mcppinchtab/pinchtab | Skill | 11k | 10d ago | MIT | A | 52 | Claude Code, Codex |
| azure-messaging-webpubsub-javamicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 43 | Claude Code |
| cua-driverdavidondrej/skills | Skill | 4.3k | 3d ago | MIT | A | 68 | Claude Code, Codex |
| open-design-browsercompozy/compozy | Skill | 2.8k | today | MIT | A | 56 | Claude Code, Codex |
| southwestborski/travel-hacking-toolkit | Skill | 895 | today | MIT | A | 40 | Claude Code |