unbrowse agent skill by unbrowse-ai
Searches and calls websites through Unbrowse's API or remote MCP, reusing indexed site tools, reading pages, and learning missing routes in its cloud browser.
- Publisher
- unbrowse-ai
- License
- MIT
- Installs via skills.sh
- 666
- Bundled scripts
- None
- Credentials
- UNBROWSE_API_KEY
- Outside hosts
- None
- MCP server
- Yes
- Security scan
- Not scanned by the directory. Third-party audits on skills.sh
- Listed since
- October 19, 2026 or earlier
At a glance
| Publisher | unbrowse-ai |
|---|---|
| Repository | github.com/unbrowse-ai/unbrowse |
| Category | AI Agents and LLM Engineering |
| Topics | MCP Servers, Browser Agents and Computer Use |
| Cost | Free skill, needs a service |
| Capabilities | Quality badge, Bundles reference files, Uses an MCP server, Needs an API key or account, Runs commands |
| License | MIT |
| Last commit | October 18, 2026 (repository) |
| Listed since | October 19, 2026 or earlier |
| SKILL.md | 147 lines, 2,441 words |
| Bundled files | 2 files, including references |
| Quality score | 83/100 · B Passes all seven graded checks |
| Security scan | Not scanned |
| README badge | Get the badges for this listing |
Before you install
Read from the skill's files at the directory's October 22, 2026 snapshot: what ships with it, what its code reads, and the terms it is published under.
- Bundled scripts: none.
- Commands in SKILL.md: its code blocks run
npx, which fetches or runs software from outside the skill folder. - Credential names SKILL.md mentions:
UNBROWSE_API_KEY. - MCP: SKILL.md mentions an MCP server.
- License: MIT, an open-source or open-content license.
Install unbrowse
The skills CLI works with Claude Code, Codex, Cursor and other agents; the other routes copy the folder by hand.
npx skills add https://github.com/unbrowse-ai/unbrowse/tree/main/plugins/claude-code/skills/unbrowse
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/unbrowse-ai/unbrowse.git "$tmp/repo"
mkdir -p .claude/skills
cp -R "$tmp/repo/"plugins/claude-code/skills/unbrowse .claude/skills/
rm -rf "$tmp"
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/unbrowse-ai/unbrowse.git "$tmp/repo"
mkdir -p .agents/skills
cp -R "$tmp/repo/"plugins/claude-code/skills/unbrowse .agents/skills/
rm -rf "$tmp"
For a user-level install, copy the folder to ~/.claude/skills/ (Claude Code) or ~/.agents/skills/ (Codex). Inside Codex, $skill-installer can also fetch a skill from GitHub. New to skills? Read How to install agent skills in Claude Code, Codex and Cursor.
unbrowse-ai/unbrowse is third-party code. Read SKILL.md and any bundled scripts before you install it.
Directory summary
unbrowse searches and calls websites through Unbrowse's API or remote MCP, reusing indexed site tools, reading pages, and learning missing routes in its cloud browser. Listed in AI Agents and LLM Engineering, it is built for Claude Code users working on AI agent and LLM engineering. It scores 83/100, grade B, on the Suede Skill Check, with full points for structural checks, recent maintenance and instruction depth. unbrowse-ai publishes it under MIT, an open-source or open-content license, so you can use and adapt it at no cost under those terms. It installs in Claude Code, Codex, Cursor and other agents that read the Agent Skills format.
How the author describes it
Search and call websites through Unbrowse's hosted API or remote MCP, reuse indexed site tools, read pages, and learn missing routes in its cloud browser. Use for structured website tasks, and authenticated site access with Unbrowse.
From the description field of SKILL.md. Agents read this line to decide when to load the skill.
Sections in SKILL.md
- Onboarding: sign in with OAuth
- Host plugins
- Choose the interface
- Execute a task
- Unbrowse a URL
- A site's tools over plain HTTP
- Read or learn a site
- Bring the agent's MCP servers
- Website sign-in
- Serving many users (orgs)
- Limits and recovery
Files in the skill folder
SKILL.mdreferences/tools.json
Suede Skill Check: 83/100 · B
Report snapshot · · graded-v2
The rubric totals 100 points; published scores are capped at 99. Security scan: not scanned.
Source revision: e8f5f04b761a. Report version a631dffb40f5. Badge URLs follow the latest listing; save the report to keep this version.
Download report · Request a fresh review
What to review next
- Boundary clause: 0 of 4 points. 4 when the description says when not to use the skill (“Not for …”, “Do not use …”, “Use … instead”).
- Example request: 0 of 3 points. 3 when the description quotes an example request (3–60 characters in quotation marks).
- Bundled scripts: 0 of 2 points. 2 when the skill folder ships a scripts/ folder.
Add declarations and bundled files when they describe the skill’s actual needs. Update the source, then request a fresh review with the changed commit or PR.
unbrowse passes the required frontmatter check and all seven graded checks, which cover the SKILL.md format, a declared license and a commit in the last year. Those checks earn 56 of the score’s points; the rest come from description quality, instruction depth, bundled files, maintenance and curated-list citations. How the score works.
See the checks and points
| Check | Result |
|---|---|
| Frontmatter parses and declares a name and a description | Pass |
| Name uses lowercase letters, digits and hyphens, 64 characters or fewer | Pass |
| Name matches the skill's folder name | Pass |
| Description is between 50 and 1,024 characters | Pass |
| Description says when an agent should use the skill | Pass |
| SKILL.md body is 500 lines or fewer | Pass |
| A license is declared in the frontmatter or the repository | Pass |
| Repository has a commit in the last 12 months | Pass |
| Score component | Points |
|---|---|
| Structural checks | 56 of 56 |
| Description length | 5 of 5 |
| Boundary clause | 0 of 4 |
| Example request | 0 of 3 |
| Instruction depth | 7 of 7 |
| Reference files | 3 of 3 |
| Bundled scripts | 0 of 2 |
| Recent maintenance | 9 of 9 |
| Curated-list citations | 3 of 7 |
| Tool permissions declared | 0 of 2 |
| Compatibility declared | 0 of 2 |
Where this skill is cited
unbrowse-ai/unbrowse appears in 2 curated lists: VoltAgent/awesome-agent-skills, heilcheng/awesome-agent-skills.
Strengths and open points
What it does well
- Full points for structural checks (56 of 56).
- Full points for recent maintenance (9 of 9).
- Full points for instruction depth (7 of 7).
- Full points for description length (5 of 5).
- Full points for reference files (3 of 3).
- Its repository is cited by 2 curated lists.
What to check
- Boundary clause: 0 of 4 points.
- Example request: 0 of 3 points.
- Bundled scripts: 0 of 2 points.
- Tool permissions declared: 0 of 2 points.
- Compatibility declared: 0 of 2 points.
- Needs UNBROWSE_API_KEY set in your environment.
- Connects to an MCP server you configure separately.
- Security scan: not scanned by the directory. Read SKILL.md and any scripts before you install.
Reviews and ratings
Directory rating: 83/100 · B on the Suede Skill Check, computed from the skill's published files.
Used unbrowse? Reader reviews are published on GitHub as issues credited to their authors. They never change the score.
Questions and answers
What does unbrowse do?
Searches and calls websites through Unbrowse's API or remote MCP, reusing indexed site tools, reading pages, and learning missing routes in its cloud browser. It is listed in AI Agents and LLM Engineering on SuedeAI.directory.
Is unbrowse free?
The skill is free under MIT. Its files reads UNBROWSE_API_KEY and mentions an MCP server, so the service it talks to may need an account or a paid plan.
How do I install unbrowse?
Run npx skills add https://github.com/unbrowse-ai/unbrowse/tree/main/plugins/claude-code/skills/unbrowse from your project, or copy the folder into .claude/skills/ for Claude Code or .agents/skills/ for Codex. The install section lists each route.
Does unbrowse need an API key or MCP server?
Yes. At the October 22, 2026 snapshot its files reads UNBROWSE_API_KEY and mentions an MCP server. The Before you install section lists the details.
How good is unbrowse?
It scores 83/100, grade B, on the Suede Skill Check, which adds structure, description quality, instruction depth, bundled files, maintenance and curated-list citations from its published files.
What are alternatives to unbrowse?
Skills that share its topics include add-webmcp from browserbase/skills, mcp-builder from anthropics/skills and tool-design from muratcankoylan/Agent-Skills-for-Context-Engineering.
Similar skills
Skills from other publishers that share the most topics with unbrowse.
Read next
More AI agent skills
Two badges for unbrowse
Choose the numeric badge, the letter badge, or both. Each links to this skill’s Suede AI listing; either valid linked badge can qualify for the feature offer.
[](https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/)
[](https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/)<a href="https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/"><img src="https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/score.svg" alt="Suede Skill Check: 83/100" width="160" height="175"></a>
<a href="https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/"><img src="https://skills.suedeai.ai/directory/unbrowse-ai/unbrowse/unbrowse/badge.svg" alt="Suede Skill Check: Grade B" width="160" height="175"></a>Suede Certified on the shield refers to this published structural quality check. Security scan: not scanned. Read the methodology.
Limited-time feature offer: Display either quality badge in your repository or skill README, linked to your listing, then request verification. Suede AI checks the rendered README at a recorded commit. Eligible skills scoring 80–99 with a verified badge can receive a featured placement while the offer is open. Request badge verification.
Publishers: read How to get your agent skill found in search and cited by AI assistants.