operations / skill

UiPath Agent Skills

Skills for AI coding agents that cover UiPath work, from RPA workflows and agents to platform operations and troubleshooting.

Repository owner UiPath

Not rated yet
0 comments

Like 0 Save

Published 11 Oct 2026Listing updated 12 Oct 2026

At a glance

Purpose: Skills for AI coding agents that cover UiPath work, from RPA workflows and agents to platform operations and troubleshooting.
Runs in: Local agent with shell
Operating systems: macOS, Linux, Windows
Needs: Runtime: Node.js >=20, UiPath CLI, .NET SDK 8, Python 3.11-3.14; Needs internet: Yes; Needs accounts or API keys beyond your agent: Not stated
Files and data: Reads files: Yes; Writes files: Yes; Sends data to other services: Yes
Tool type: Skill
Delivery: From GitHub
Version: Latest on the main branch
Access: Code on GitHub
Compatibility: According to the README, the repository works with Claude Code, Google Gemini CLI, OpenAI Codex CLI and Cursor IDE. The README describes a Claude Code plugin. The README describes AGENTS.md, .agents/skills/ and Codex plugin metadata for Codex CLI, and requires Codex CLI 0.142.0 or later. The README says Cursor automatically loads project rules from .cursor/rules/. The README gives separate installer commands for macOS/Linux and Windows.
External costs: The repository is MIT-licensed and states no pricing. Your coding agent or model provider may charge for usage. The README does not say what UiPath services cost.
Permissions: The skills act through your coding agent with the access you give it. The README says uip skills install keeps skills updated automatically. The README describes account authentication, automation authoring, execution, publishing, deployment, schedules, platform administration and feedback submission through uip. The README's Windows troubleshooting includes git checkout -- ., which can discard local edits to tracked files, and enabling Git long-path support through a global configuration change. The README says Claude Code prompts for each uip command by default. The README says /uipath:install-permissions offers to merge an allowlist into ~/.claude/settings.json or .claude/settings.local.json and notes a session-start reminder and shared session hooks. The README describes automatic Project Discovery writes to .claude/rules/project-context.md and AGENTS.md.
Usage terms: MIT License, as stated in the repository. See its LICENSE file for the exact terms.

Code on GitHub

UiPath/skills

MIT

Python168 stars59 forksLast pushed 13 Oct 2026 UTC

This is a repository of skills for interfacing UiPath capabilities to external developers.

Public GitHub data fetched 13 Oct 2026 UTC · 0 days old

View on GitHub ↗

What it does

UiPath Agent Skills is a repository of skills for AI coding agents. The README says the skills provide domain knowledge for building, running, testing and deploying UiPath automations and agents. The README describes each skill as a self-contained package of instructions and resources for a UiPath task.

The README groups its catalog into Authoring, Solution & Planning, Platform & Operations, and Diagnostics & Feedback. The README lists authoring skills for RPA, Maestro, agents, coded apps, API workflows, human tasks and document understanding. The README lists planning and solution skills, a uipath-review skill described as a read-only auditor, platform and administration skills, Test Manager work, troubleshooting and feedback submission. The README also describes a Project Discovery agent that writes context files automatically when an eligible UiPath project lacks them, or on explicit request.

The README labels the repository Work in Progress and marks skill maturity as Stable, Preview or In-development. The README names Claude Code, Google Gemini CLI, OpenAI Codex CLI and Cursor IDE as supported tools.

Who it is for

Developers who use an AI coding agent to work with UiPath automations, agents and platform services.

Example tasks

The README lists these tasks:
Create and debug RPA workflows with uipath-rpa.
Turn a Process Design Document into a Solution Design Document with uipath-planner.
Scaffold, build and publish coded apps with uipath-coded-apps.
Investigate failures and stuck jobs with uipath-troubleshoot.
Manage Test Manager projects, cases and sets with uipath-test.

Getting started

The README's quick start downloads and executes an onboarding script through curl piped to bash on macOS/Linux or irm piped to iex on Windows. The README says the installer runs uip skills install --no-interactive to wire skills into detected coding agents. To rerun installation or choose an agent, the README gives uip skills install and --agent <name>. For Codex CLI, the README gives codex plugin marketplace add UiPath/skills --ref main followed by codex plugin add uipath@uipath-marketplace. For a single skill, the README shows npx skills add UiPath/skills --skill uipath-rpa. The README prefers uip skills install for ongoing use and advises against manual Claude Code plugin installation. The README has the exact commands.

Limitations

According to the README, the repository is a work in progress. It marks each skill as Stable, Preview or In-development. It says Preview skills are not yet stable and may be gated or allowlisted, and that In-development skills are incomplete or unstable. The README says the quick-start installer installs Node.js 20 or later, @uipath/cli, .NET SDK 8.0 and Python 3.11 to 3.14 along with the skills. The README requires Codex CLI 0.142.0 or later for its Codex route. The README tells Windows users to clone with git symlinks enabled. The README says Claude Code prompts for approval on every uip command by default. The repository can change at any time; check its README before you install.

Data practices

The README says Project Discovery writes project context to .claude/rules/project-context.md and AGENTS.md. The README describes platform work involving Orchestrator, Data Fabric, LLM Gateway and traces, and says uipath-feedback submits bug reports and improvement suggestions through uip feedback send. The README does not describe telemetry or retention. The top-level entries include TELEMETRY.md and SECURITY.md; check those files for details. What your coding agent and model provider receive depends on how you use the skills.