agentic-browser
Automate Browser Workflows with Agentic Browser
Manual browser tasks slow research, testing, and data collection. This skill lets Claude, Codex, and Claude Code operate a browser through inference.sh sessions.
Do not auto-install this skill.
The canonical policy requires operator review before any installation action.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "agentic-browser" from https://skillstore.io/skills/inference-sh-agentic-browser.md and its manifest at https://skillstore.io/api/skills/inference-sh-agentic-browser/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "agentic-browser". Open a product page and identify checkout controls.
Expected outcome:
The page loaded successfully. I found product options, quantity controls, an add-to-cart button, and a checkout link.
Using "agentic-browser". Capture the current page after submitting a contact form.
Expected outcome:
The form was submitted. The new page state shows a confirmation message and no visible validation errors.
Using "agentic-browser". Extract headings from a documentation page.
Expected outcome:
The page contains setup, authentication, sessions, functions, examples, and troubleshooting sections.
Security Audit
CriticalThe reviewed skill is primarily documentation for controlling an inference.sh browser session. Most static findings are false positives from markdown code fences and inline backticks, but the quick-start command that pipes a remote installer into sh is a confirmed critical supply-chain risk.
Confirmed security concerns (1)
Capability review items (1)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Risk Factors
⚙️ External commands (51)
🌐 Network access (9)
Detected Patterns
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/inference-sh-agentic-browser/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/inference-sh-agentic-browser?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/inference-sh-agentic-browser?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/inference-sh-agentic-browser/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/inference-sh-agentic-browser.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
inference-sh. (2026). agentic-browser security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/inference-sh-agentic-browser/audits/4BibTeX citation
@techreport{inference-sh-inference-sh-agentic-browser-2026,
author = {inference-sh},
title = {agentic-browser security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/inference-sh-agentic-browser/audits/4},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "agentic-browser security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "inference-sh"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/inference-sh-agentic-browser/audits/4"
identifiers:
- type: other
value: "skillstore:inference-sh-agentic-browser:audit:4"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: MediumWhat You Can Build
Test Web Form Flows
Open a form, fill fields, click controls, and capture the resulting page state.
Collect Research Evidence
Navigate pages, inspect visible elements, and capture screenshots for review notes.
Extract Website Data
Use page snapshots and controlled JavaScript to collect repeated data from authorized sites.
Try These Prompts
Use agentic-browser to open this URL: [URL]. Report the page title, current URL, and main interactive elements.
Open [URL], identify the form fields, fill them with these values: [VALUES], then show the resulting page state.
Navigate through this workflow on [URL]: [STEPS]. After each navigation, refresh the snapshot and note visible success or error states.
Open [URL], inspect the page structure, run safe page JavaScript to collect [DATA], and summarize the extracted results.
Best Practices
- Start with a new session, then reuse the returned session ID for related actions.
- Refresh the snapshot after navigation, form submission, or dynamic page updates.
- Close browser sessions when the task is complete.
Avoid
- Do not reuse element references after the page changes or navigates.
- Do not automate sites where you lack permission to browse or submit data.
- Do not place passwords or sensitive tokens into prompts unless the site is trusted.
Frequently Asked Questions
What does this skill control?
Does it require inference.sh?
Can it click and fill forms?
Can it run JavaScript?
When should I refresh the snapshot?
Is the quick-start installer safe?
Developer Details
Author
inference-shLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
a06681402992ceae98ba04d54cfd4ab004862696
Maintenance freshness
7/21/2026
Usage
6 downloads · 219 views
File structure
📄 SKILL.md