Maintainer of this skill?
by openyak
One chat. Every agent. A universal interface for AI agents running on your computer.
# Add to your Claude Code skills
git clone https://github.com/openyak/openyakLast scanned: 5/5/2026
{
"issues": [
{
"type": "npm-audit",
"message": "axios: Axios has a NO_PROXY Hostname Normalization Bypass that Leads to SSRF",
"severity": "high"
},
{
"type": "npm-audit",
"message": "follow-redirects: follow-redirects leaks Custom Authentication Headers to Cross-Domain Redirect Targets",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "lodash: lodash vulnerable to Code Injection via `_.template` imports key names",
"severity": "high"
}
],
"status": "WARNING",
"scannedAt": "2026-05-05T06:19:42.234Z",
"semgrepRan": false,
"npmAuditRan": true,
"pipAuditRan": true
}See how openyak compares with popular alternatives.
openyak is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by openyak. One chat. Every agent. A universal interface for AI agents running on your computer. It has 692 GitHub stars.
openyak returned warnings in SkillsLLM's automated security scan. It has no critical vulnerabilities, but review the flagged issues in the Security Report section before adding it to your workflow.
Clone the repository with "git clone https://github.com/openyak/openyak" and add it to your Claude Code skills directory (see the Installation section above).
openyak is primarily written in TypeScript. It is open-source under openyak on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh openyak against similar tools.
No comments yet. Be the first to share your thoughts!
Requires a passing catalog security scan.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Your agents. Your conversation. One desktop workspace.
OpenYak is a local desktop GUI for Codex and Claude Code. Keep the conversation, inspect the files, and work alongside your agent in a shared browser—without making the terminal your workspace.
This is v2, on
main, and still alpha. The original v1 is preserved onlegacy/v1and in the v1.5.0 release. OpenYak integrates supported runtime interfaces; it does not reproduce every private feature of Codex Desktop or Claude Desktop.
Real OpenYak v2, dark mode. The fictional Orbit project is isolated demo data; Codex generated the report during capture.
Open file references directly from a response. Read Markdown as a document, preview sandboxed HTML, view PDFs and DOCX documents, or inspect code with highlighting and line numbers. Opening another file keeps existing tabs available.
The runtime boundary normalizes supported structured artifact and file outputs; the frontend does not need to guess tool names or turn arbitrary code blocks into artifacts. Ordinary file references are resolved through a separate file-opening path.
The dashboard above is a hand-authored fixture rendered by the real app—not a generated screenshot or a claim that the agent built the dashboard.
Use your existing provider authentication. Choose the models and session options the runtime exposes, without copying private Desktop system prompts into OpenYak.
| Provider | Default integration | Optional compatibility path |
|---|---|---|
| Codex | Native App Server over stdio | Codex ACP adapter |
| Claude Code | Claude Agent SDK driving its CLI | Claude ACP adapter |
Model availability, usage limits, and provider features depend on your account and installed runtime versions. OpenYak does not supply model access.
The shared browser uses Playwright MCP and a dedicated Chrome session. The agent and the user operate the same page. Taking control blocks new agent browser actions and waits for in-flight work; resuming returns browser access without sending another chat prompt.
The panel uses lossless HiDPI frames. It is a shared remote view, not an embedded native browser or a promise of 60 fps. The pictured flow was exercised through Codex in the real GUI. External computer-use tools and native desktop control are separate capabilities; this screenshot does not certify those integrations.
Read the shared-browser architecture and boundaries.
Prerequisites: Node 26, Rust 1.90 (mise install installs the pinned toolchains), and an existing Codex and/or Claude Code sign-in. Install Google Chrome to use the shared browser.
git clone https://github.com/openyak/openyak.git
cd openyak
npm install
npm run dev
This builds the Rust core and launches Electron with hot reload. Agent dependencies are pinned in the app; authentication remains yours. OPENYAK_CODEX_BIN and OPENYAK_CLAUDE_BIN can select compatible local CLI binaries—see the runtime documentation.
For an isolated test instance, set OPENYAK_DATA_DIR to a new, empty directory. To opt into ACP, use OPENYAK_AGENT_TRANSPORT=acp; native host integrations such as the shared browser are not automatically available on that path.
app/ Electron + React — chat, file workbench, native workers, host integrations
core/ Rust + SQLite — projects, tasks, transcripts, normalized runtime events
docs/ — architecture, contracts, and integration boundaries
OpenYak stores its conversation data locally. That does not mean offline inference: provider requests and browser navigation can use the network.
See the native runtime design, app/core protocol, and screenshot provenance and reproduction guide.
For logos, banners and visual guidelines, start with the OpenYak brandkit.
Packaged installers, broader agent support, and Linux/Windows GUI validation are still work in progress. Runtime-reported subagents are not a complete workflow-orchestration UI. Claude integration exists, but the browser screenshots here validate Codex, not a Claude browser acceptance run. Private Desktop-only tools and full computer-use parity are not guaranteed.
Try a real task, switch agents, inspect the output, and tell us what breaks. See CONTRIBUTING.md.
npm run check
npm run build
The v2 CI workflow checks runtime contracts, production builds, and real desktop/browser regressions without paid model calls. Documentation-only changes skip the heavy checks.