https://retellai.com/?utm_source=discord&utm_medium=seo logo
how can i make an agent check BST time zone
# ❓・ai-evy-support
k
@AI Evy how can i make an agent check BST time zone?
a
Thanks for tagging me @K Rudy. I'll look through relevant sources and get back to you shortly.
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
### Answer To make your agent check BST (British Summer Time) time zone, you can use the timezone-specific dynamic variable syntax. [(1) ]() Use
{{current_time_Europe/London}}
in your prompts, begin message, or tool configurations. This will provide the current time in the Europe/London timezone, which observes BST during summer months. For example, in your agent prompt:
Copy code
Now is {{current_time_Europe/London}}. Use this time for all scheduling and time-based decisions.
You can also use
{{current_hour_Europe/London}}
to get the current hour as a fraction, or
{{current_calendar_Europe/London}}
for a 14-day calendar in that timezone. If you need to use this in a Logic Split node for time-based routing, you can create equation-based conditions like:
Copy code
{{current_hour_Europe/London}} >= 9 AND {{current_hour_Europe/London}} < 17
[(2) ]() This approach is more reliable than letting the LLM interpret time from prompts alone.
​ To learn more, see the sources I considered: [[1] Dynamic Variables](https://docs.retellai.com/build/dynamic-variables) [[2] Agent not following time-based logic split](https://discord.com/channels/1186424376300679209/1465785079111483464) ​
k
@AI Evy coould you tell me mroe about that logic split node, is that avilabble on prompt bases agents? or only in conversation flow?
a
Thanks for the follow-up, @K Rudy. I'll get back to you soon.
### Answer Logic Split nodes are only available in Conversation Flow agents, not in prompt-based agents. [(1) ]() [(2) ]() [(3) ]() The Logic Split node is a specific node type used to branch conversation flow based on conditions. [(1) ]() When the agent enters this node, it immediately evaluates conditions and branches to the corresponding destination nodes without speaking. Retell AI offers different agent types: [(2) ]() [(3) ]() - **Conversation Flow Agent**: Uses node-based editor with Logic Split nodes, Function nodes, Call Transfer nodes, etc. - **Single Prompt Agent**: Uses one comprehensive prompt for simple use cases [(2) ]() - **Multi-Prompt Agent**: Combines several prompts in a structured tree but uses transition logic between nodes For prompt-based agents (Single Prompt or Multi-Prompt) , you would need to handle time-based logic through the agent's prompt instructions rather than using Logic Split nodes. The visual node-based builder with Logic Split functionality is exclusive to Conversation Flow agents. [(1) ]() [(2) ]() [(3) ]()
​ To learn more, see the sources I considered: [[1] Logic Split Node](https://docs.retellai.com/build/conversation-flow/logic-split-node) [[2] Retell AI’s Advanced Conversation Flow | Retell AI](https://www.retellai.com/blog/unlocking-complex-interactions-with-retell-ais-conversation-flow) [[3] Inside Retell AI | The Conversational AI Platform Behind Scalable AI Phone Systems | Retell AI](https://www.retellai.com/blog/inside-retell-ai-conversational-ai-phone-system) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
k
@AI Evy it’s currently a prompt based one, is it reliable to use there!
a
Thanks for the follow-up, @K Rudy. I'll get back to you soon.
### Answer Using time-based logic in prompt-based agents is less reliable than in Conversation Flow agents. [(1) ]() For prompt-based agents, you should include the timezone variable in your agent prompt and give explicit instructions about when to use tools or take actions. For example, you can add to your prompt: [(2) ]()
Copy code
You have access to the current time: {{current_time_Europe/London}}. You can tell the user the current time if asked.
However, the Prompt Engineering Guide specifically recommends switching to Conversation Flow when you need reliable time-based decision branches. The guide states that you should use Conversation Flow agents when dealing with "multiple decision branches" or when "reliability concerns" arise where "single prompt shows inconsistent behavior." [(1) ]() From Discord community experiences, prompt-based logic for time checks works but can be unreliable - one user reported that "prompt based agents have a terrible time with determining whether or not it is inside business hours" and typically had to write custom functions to handle it programmatically. [(2) ]() If you need deterministic time-based routing (like business hours vs. after hours) , Conversation Flow with Logic Split nodes will be more reliable than relying on the LLM to interpret time conditions in a prompt. [(1) ]()
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
​ To learn more, see the sources I considered: [[1] Prompt Engineering Guide](https://docs.retellai.com/build/prompt-engineering-guide) [[2] Agent not following time-based logic split](https://discord.com/channels/1186424376300679209/1465785079111483464) ​
k
@AI Evy What about current_time itself, is that reliable on the date and time it gives?
a
Thanks for the follow-up, @K Rudy. I'll get back to you soon.
### Answer Yes,
{{current_time}}
is a reliable default system variable that Retell automatically provides. [(1) ]() It returns the current time in
America/Los_Angeles
timezone. For example, the format is: "Thursday, March 28, 2024 at 11:46:04 PM PST". For BST specifically, you should use
{{current_time_Europe/London}}
instead, which will give you the accurate time in that timezone. The timezone-specific variables follow the pattern
{{current_time_[timezone]}}
. In practice, a workshop tutorial demonstrates using this variable by including it in the agent context like this: [(2) ](

▾

)
Copy code
The current date and time is {{current_time}}
The dynamic variable itself is reliable and automatically updated by Retell's system. [(1) ]() However, as discussed in the previous messages, how the LLM interprets and acts on that time information in prompt-based agents can be less reliable than using equation-based Logic Split nodes in Conversation Flow. [(3) ]()
​ To learn more, see the sources I considered: [[1] Dynamic Variables](https://docs.retellai.com/build/dynamic-variables) [[2] Part 1: BuildYour First Retell Voice Al Agent That Sells Cars](

https://www.youtube.com/watch?v=p0DusrDjfjA▾

) [[3] Agent not following time-based logic split](https://discord.com/channels/1186424376300679209/1465785079111483464) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
k
@AI Evy ok good, what about the time for denver
a
Thanks for the follow-up, @K Rudy. I'll get back to you soon.
### Answer For Denver time, use
{{current_time_America/Denver}}
. [(1) ]() Denver is in the Mountain Time Zone, which observes Mountain Standard Time (MST) or Mountain Daylight Time (MDT) depending on the season. You can use this in your agent prompt, begin message, or tool configurations just like the BST example:
Copy code
Now is {{current_time_America/Denver}}. Use this time for all scheduling and time-based decisions.
You can also use
{{current_hour_America/Denver}}
to get the current hour as a fraction, or
{{current_calendar_America/Denver}}
for a 14-day calendar in the Denver timezone.
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
​ To learn more, see the sources I considered: [[1] Dynamic Variables](https://docs.retellai.com/build/dynamic-variables) ​
k
hi @AI Evy how can i make the inbound webhook of a call send dynamic variables?, whats the strcutura i need to respond with?
a
Thanks for the follow-up, @K Rudy. I'll get back to you soon.
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
### Answer For the Inbound Call Webhook, you need to respond with a JSON object containing key-value pairs of the variables you want to access in your prompt. [(1) ]() The structure should be:
Copy code
json
{
  "first_name": "John",
  "email": "john@example.com",
  "phone_number": "+18005551212"
}
These variables will then be available in your agent prompt using the
{{variable_name}}
syntax. For example,
{{first_name}}
,
{{email}}
, and
{{phone_number}}
. Important: Do NOT wrap your variables in a
retell_llm_dynamic_variables
object for the Inbound Call Webhook response. That structure is only used when making outbound calls via the API. The Inbound Call Webhook expects the variables directly as top-level key-value pairs. If your dynamic variables aren't populating, double-check that: - Your webhook returns a 200 status code [(2) ]() - The JSON structure matches the format above (no nested objects) [(1) ]() [(2) ]() - The variable names in your prompt match exactly what you're returning from the webhook [(3) ]()
​ To learn more, see the sources I considered: [[1] Discord message from @fyodor2572 on 2025-01-13](https://discord.com/channels/1186424376300679209/1186462578310910112/1328501089112424508) [[2] Dynamic variables not populating](https://discord.com/channels/1186424376300679209/1346952882091720855) [[3] Dynamic Variables](https://docs.retellai.com/build/dynamic-variables) ​