Table of contents
Official Content
  • This documentation is valid for:

Public Tools that do not require additional configuration are available by default in all Projects.

You can add them directly from the AI & Tools Tab of an Agent, without having to create a Secret or complete additional parameters. To do this, go to the Tools and Agents section and select + Add to open the tool selection dialog.

PublicToolsConfiguration

In that dialog, you can apply a filter to show only Public Tools. Some tools display a warning icon, indicating that they require configuration before they can be used. Tools without that icon can be added directly to the Agent.

The following Public Tools do not require configuration.

arxiv_search

com.globant.geai.arxiv_search: Searches for research papers on arXiv.org, an open-access repository for fields such as physics, mathematics, computer science, and more.

chat_with_rag

com.globant.geai.chat_with_rag: Allows an Agent to interact directly with one or more RAG Assistants configured in the same Project. For more information, see RAG Tool.

create_image

com.globant.geai.create_image: Creates images based on natural language descriptions using AI models. Just type what you want, and the Tool will generate an original image.

You can configure the model to use by setting the provider/model value. This configuration can be defined:

  • In the Agent Guidelines, within the Agent Setup section.
  • Or, allow the consumer to specify the desired model at runtime.

If no model is specified, openai/gpt-image-1 is used by default.

The models you can configure are as follows:

  • openai/gpt-image-1
  • openai/dall-e-3
  • vertex_ai/imagen-3.0-generate-001
  • vertex_ai/imagen-3.0-fast-generate-001
  • vertex_ai/imagen-3.0-generate-002
  • xai/grok-2-image-1212

Once generated, the Tool returns the URL of the resulting image.

pubmed_fetch

com.globant.geai.pubmed_fetch: Retrieves detailed information about research papers in PubMed using a list of paper IDs (PMIDs). It provides key details such as title, authors, abstract, publication date, and journal information. It is recommended to use it with com.globant.geai.pubmed_search.

pubmed_search

com.globant.geai.pubmed_search: Searches for research papers in PubMed, the leading database for biomedical and life sciences literature. It returns a list of paper IDs (PMIDs) that match your query, which you can then use to fetch detailed information. It is recommended to use it in conjunction with com.globant.geai.pubmed_fetch.

web_scraper_httpx

com.globant.geai.web_scraper_httpx: Extracts information from websites. Instead of manually copying data, this Tool fetches it for you and presents it in an organized format. You must provide a URL or you can use it in conjunction with com.globant.geai.web_search.

web_search

com.globant.geai.web_search: Searches the Internet for real-time information. It is useful for accessing current news, discovering events, finding local information, researching topics, and more. It is recommended to use it in conjunction with com.globant.geai.web_scraper_httpx.

Tools for use in an Agentic Process

The following tools are used when the Agent is running within an Agentic Process.

create_artifact

com.globant.geai.create_artifact: This Tool is used in an Agentic Process. Writes generated content into an artifact (file) and returns the file path where the artifact was stored.

get_user_feedback

com.globant.geai.get_user_feedback: Requests feedback from the user about your work. This Tool is used only in an Agentic Process.

read_process_artifact

com.globant.geai.read_process_artifact: This Tool is used in an Agentic Process. Reads the content of an artifact (file). Files can be PDFs of plain text.

Last update: December 2025 | © GeneXus. All rights reserved. GeneXus Powered by Globant