Activity
Mon
Wed
Fri
Sun
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
What is this?
Less
More

Memberships

FM
FBA Mastery Syndicate

610 members • Free

AI Automation (A-Z)

99.7k members • Free

Agent Zero

1.6k members • Free

AI Developer Accelerator

9.9k members • Free

6 contributions to AI Developer Accelerator
The root agent gets out of the loop.
I have declared a root_agent which is supposed to follow the workflow but after triggering the first agent, it stops there. Below is the instruction of my root_agent. You are a manager of a care home. Your primary task is to help the staff member creating an event by analyzing and classifying the description provided by staff. Your workflow: 1. **Validate the event description**: Ensure that the description is clear and contains all necessary information (e.g., name or initials of resident/staff/manager, date, location (optional), etc.). If the description is unclear or missing critical information, provide feedback to the user. To do this use the `event_validator` tool. 2. **Extract relevant information**: If the event description is valid, extract key details such as the eventType(can be `resident` or `staff`) and names of residents or staff mentioned. Use the `event_information_extractor` tool for this task. If no resident or staff is found, return an error message to the user. 3. **Classify the event**: Based on the extracted information, classify the event using the `event_classifier` tool.
0 likes • 1d
Try this:- You are a manager of a care home. Follow this exact workflow in sequence. Do not stop after the first step always continue until all steps are executed. Workflow: 1. Call the `event_validator` tool to check description validity. - If invalid --> return feedback to the user. - If valid --> continue to step 2. 2. Call the `event_information_extractor` tool to extract eventType and names. - If no staff/resident found --> return error to user. - If found --> continue to step 3. 3. Call the `event_classifier` tool to classify the event.
Subagents as tool not able to call custom tools.
I have created Agent system with 3 subagents as tool. I have cleary given instructions to a subagent to use a tool to get the information from the tool but it's not working.
0 likes • 1d
In your ADK system config, make sure the custom tool is registered under the same subagent.
0 likes • 1d
Instead of just: Use the tool to fetch finance data. Use something like: To answer, you must call the fetch_reports tool with the appropriate parameters.
Hey amazing people! 😄
Just landed here thanks to a YouTube course and thought I’d say hi! Super pumped to be part of this supportive community everyone seems so welcoming already 🙌 I’m here to learn, grow, and hopefully give back as I go. If anyone has tips for beginners, I’m all ears! 👂💬 Let’s connect, share, and learn together on this journey! 🚀❤️
0 likes • 4d
@Fitz A Tnq Fitz I hope I will learn from you all 🪯
0 likes • 2d
@Tom Welsh Thanks so much🙂, Tom! Really happy to be here. Right now I’m diving into some AI projects mainly around agent building, cloud integrations, automation, and a bit of the machine learning workflow space. I’ll do my best to join the Tuesday calls live, and it’s awesome that replays are available if I can’t make it. Excited to learn, share, and contribute with everyone here 👋
Help Needed: ADK OneDrive Custom Integration
Hi everyone 👋 I’m currently working on building a OneDrive connector using ADK. Has anyone here worked on something similar or has experience with it? Any guidance or references would be really helpful!
0 likes • 4d
Anyone is familiar with Agent-Space or integration connectors #all
How to get back to main agent?
I have created an Agent system using ADK which has one manager which is managing other agents as sub agents. So, when the manager delegates to agent A, i want to return to the main agent and then continue the process.
2 likes • 4d
In the prompt, whenever you are developing a sub-agent, clearly specify that if the sub-agent does not have the required data or cannot handle the condition, it should always return control to the root (main) agent. This fallback to the root agent should be explicitly mentioned in the query or defined as a return-to-root rule @Jahangir Jadi
1-6 of 6
Purendeeswar Reddy
2
15points to level up
@purendeeswar-reddy-6841
Agent

Active 18h ago
Joined Sep 19, 2025
India
Powered by