메뉴
Natural language web UI control — element detection, targeted interaction, and automated form filling
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/jeremylongshore/oss-agent-lab --skill gui-agent명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
SKILL.md 표시 중
| name | gui_agent |
| display_name | GUI Agent Specialist |
| description | Natural language web UI control — element detection, targeted interaction, and automated form filling |
| version | 0.1.0 |
| source_repo | alibaba/page-agent |
| license | Apache-2.0 |
| tier | core |
| capabilities | ["gui_automation","web_interaction","element_detection","form_filling"] |
| allowed_tools | ["detect_elements","interact_element","fill_form"] |
| output_formats | ["python_api","cli","mcp_server","agent_skill","rest_api"] |
The GUI Agent specialist wraps the alibaba/page-agent natural-language web UI control framework into a composable OSS Agent Lab specialist. Given a plain-English instruction and a target URL, it:
Each stage is independently callable as a tool, making the specialist suitable for targeted automation steps as well as full end-to-end web-interaction pipelines.
| Tool | Description | Parameters | Side Effects |
|---|---|---|---|
detect_elements | Detect interactive UI elements on a page | url, description | None (read-only) |
interact_element | Perform an action on a specific element | element_id, action, value | Modifies page state |
fill_form | Fill a web form with field-value pairs | url, form_data | Modifies page state |
detect_elements
url: str — fully-qualified URL of the target page (required)description: str | None — semantic hint to filter detected elements (default: )Noneinteract_element
element_id: str — stable element ID from detect_elements (required)action: str — "click" | "type" | "hover" | "focus" | "clear" (default: "click")value: str | None — text to type; required when action="type" (default: None)fill_form
url: str — fully-qualified URL of the page hosting the form (required)form_data: dict[str, str] — mapping of field labels to values (required)SpecialistRequest
│
▼
detect_elements(url, description)
│
▼
interact_element(element_id, action, value?)
│
▼ (when form_data in parameters)
fill_form(url, form_data)
│
▼
SpecialistResponse(result={detection, interaction, form?})
Form filling is triggered when the form_data key is present and non-empty in
request.intent.parameters.
from agents.specialists.gui_agent.agent import GuiAgentSpecialist
from oss_agent_lab.contracts import Intent, Query, SpecialistRequest
specialist = GuiAgentSpecialist()
request = SpecialistRequest(
intent=Intent(
action="click",
domain="web_interaction",
confidence=0.95,
parameters={"url": "https://example.com/login", "description": "sign in button"},
),
query=Query(user_input="click the sign in button"),
specialist_name="gui_agent",
)
response = await specialist.execute(request)
print(response.result["interaction"]["element_state"])
oss-lab run gui_agent "click the sign in button on https://example.com/login"
request = SpecialistRequest(
intent=Intent(
action="fill_form",
domain="gui_automation",
confidence=0.90,
parameters={
"url": "https://example.com/signup",
"form_data": {"email": "user@example.com", "password": "s3cr3t"},
},
),
query=Query(user_input="fill the signup form"),
specialist_name="gui_agent",
)
response = await specialist.execute(request)
print(response.result["form"]["submit_ready"])
from agents.specialists.gui_agent.tools import detect_elements
result = detect_elements(
url="https://example.com/login",
description="email field",
)
print(result["elements"])
print(f"Found {result['element_count']} element(s)")
{
"url": "https://example.com/login",
"description": "sign in button",
"detection": {
"elements": [
{"id": "elem-a1b2c3d4", "type": "button", "text": "Sign in", "selector": "button[type='submit']"}
],
"page_title": "Page at example.com — detected via page-agent (sign_in_button)",
"element_count": 1
},
"interaction": {
"success": true,
"action_performed": "click",
"element_state": "clicked",
"response_time_ms": 42.5
},
"form": {
"fields_filled": 2,
"success": true,
"validation_errors": [],
"submit_ready": true
}
}
Wraps alibaba/page-agent — a natural-language web UI control agent that translates plain-English instructions into browser automation actions using vision-language models and structured element grounding.
SOC 직업 분류 기준
