pieter/eveAI
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

- cleanup of old TASKs, AGENTs and SPECIALISTs

Browse Source 
- Add additional configuration options to agent (temperature and model choice)
- Define new PROOFREADING Agents and Tasks
This commit is contained in:
Josako
2025-10-23 09:10:52 +02:00
parent 4ec1099925
commit 1720ddfa11
27 changed files with 112 additions and 860 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
config/tasks/globals/EMAIL_LEAD_DRAFTING_TASK/1.0.0.yaml
View File

@@ -1,35 +0,0 @@
version: "1.0.0"
name: "Email Lead Draft Creation"
task_description: >
Craft a highly personalized email using the lead's name, job title, company information, and any relevant personal or
company achievements when available. The email should speak directly to the lead's interests and the needs
of their company.
This mail is the consequence of a first conversation. You have information available from that conversation in the
- SPIN-context (in between triple %)
- personal and company information (in between triple $)
Information might be missing however, as it might not be gathered in that first conversation.
Don't use any salutations or closing remarks, nor too complex sentences.
Our Company and Product:
- Company Name: {company}
- Products: {products}
- Product information: {product_information}
{customer_role}'s Identification:
$$${Identification}$$$
SPIN context:
%%%{SPIN}%%%
{custom_description}
expected_output: >
A personalized email draft that:
- Addresses the lead by name
- Acknowledges their role and company
- Highlights how {company} can meet their specific needs or interests
{customer_expected_output}
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "Email Drafting Task towards a Lead"
changes: "Initial version"

28
config/tasks/globals/EMAIL_LEAD_ENGAGEMENT_TASK/1.0.0.yaml
View File

@@ -1,28 +0,0 @@
version: "1.0.0"
name: "Email Lead Engagement Creation"
task_description: >
Review a personalized email and optimize it with strong CTAs and engagement hooks. Keep in mind that this email is
the consequence of a first conversation.
Don't use any salutations or closing remarks, nor too complex sentences. Keep it short and to the point.
Don't use any salutations or closing remarks, nor too complex sentences.
Ensure the email encourages the lead to schedule a meeting or take
another desired action immediately.
Our Company and Product:
- Company Name: {company}
- Products: {products}
- Product information: {product_information}
Engagement options:
{engagement_options}
{custom_description}
expected_output: >
An optimized email ready for sending, complete with:
- Strong CTAs
- Strategically placed engagement hooks that encourage immediate action
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "Make an Email draft more engaging"
changes: "Initial version"

24
config/tasks/globals/IDENTIFICATION_DETECTION_TASK/1.0.0.yaml
View File

@@ -1,24 +0,0 @@
version: "1.0.0"
name: "Identification Gathering"
task_description: >
You are asked to gather lead information in a conversation with a new prospect. This is information about the person
participating in the conversation, and information on the company he or she is working for. Try to be as precise as
possible.
Take into account information already gathered in the historic lead info (between triple backquotes) and add
information found in the latest reply. Also, some identification information may be given by the end user.
historic lead info:
```{historic_lead_info}```
latest reply:
{query}
identification:
{identification}
{custom_description}
expected_output: >
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task that gathers identification information from a conversation"
changes: "Initial version"

19
config/tasks/globals/IDENTIFICATION_QUESTIONS_TASK/1.0.0.yaml
View File

@@ -1,19 +0,0 @@
version: "1.0.0"
name: "Define Identification Questions"
task_description: >
Gather the identification information gathered by your team mates. Ensure no information in the historic lead
information (in between triple backquotes) and the latest reply of the user is lost.
Define questions to be asked to complete the personal and company information for the end user in the conversation.
historic lead info:
```{historic_lead_info}```
latest reply:
{query}
{custom_description}
expected_output: >
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task to define identification (person & company) questions"
changes: "Initial version"

27
config/tasks/globals/RAG_CONSOLIDATION_TASK/1.0.0.yaml
View File

@@ -1,27 +0,0 @@
version: "1.0.0"
name: "Rag Consolidation"
task_description: >
Your teams have collected answers to a user's query (in between triple backquotes), and collected additional follow-up
questions (in between triple %) to reach their goals. Ensure the answers are provided, and select a maximum of
{nr_of_questions} out of the additional questions to be asked in order not to overwhelm the user. The questions are
in no specific order, so don't just pick the first ones. Make a good mixture of different types of questions,
different topics or subjects!
Questions are to be asked when your team proposes questions. You ensure both answers and additional questions are
bundled into 1 clear communication back to the user. Use {language} for your consolidated communication.
Be sure to format your answer in markdown when appropriate. Ensure enumerations or bulleted lists are formatted as
lists in markdown.
{custom_description}
Anwers:
```{prepared_answers}```
Additional Questions:
%%%{additional_questions}%%%
expected_output: >
{custom_expected_output}
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task to consolidate questions and answers"
changes: "Initial version"

26
config/tasks/globals/RAG_PROOFREADING_TASK/1.0.0.yaml Normal file
View File

@@ -0,0 +1,26 @@
version: "1.0.0"
name: "RAG QA Task"
task_description: >
You have to improve this first draft answering the following question:
£££
{question}
£££
We want you to pay extra attention and adapt to the following requirements:
- The answer uses the following Tone of Voice: {tone_of_voice}, i.e. {tone_of_voice_context}
- The answer is adapted to the following Language Level: {language_level}, i.e. {language_level_context}
- The answer is suited to be {conversation_purpose}, i.e. {conversation_purpose_context}
- And we want the answer to have the following depth: {response_depth}, i.e. {response_depth_context}
Ensure the following {language} is used.
If there was insufficient information to answer, answer "I have insufficient information to answer this
question." and give the appropriate indication.
expected_output: >
Your answer.
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task that gives RAG-based answers"
changes: "Initial version"

34
config/tasks/globals/RAG_TASK/1.1.0.yaml
View File

@@ -7,19 +7,9 @@ task_description: >
{question}
£££
Base your answer on the following context (in between triple $):
$$$
{context}
$$$
Take into account the following history of the conversation (in between triple €):
€€€
{history}
€€€
The HUMAN parts indicate the interactions by the end user, the AI parts are your interactions.
Base your answer on the context below, in between triple '$'.
Take into account the history of the conversion , in between triple '€'. The parts in the history preceded by 'HUMAN'
indicate the interactions by the end user, the parts preceded with 'AI' are your interactions.
Best Practices are:
@@ -27,11 +17,29 @@ task_description: >
- Always focus your answer on the actual question.
- Try not to repeat your historic answers, unless absolutely necessary.
- Always be friendly and helpful for the end user.
Tune your answer with the following:
- You use the following Tone of Voice for your answer: {tone_of_voice}, i.e. {tone_of_voice_context}
- You use the following Language Level for your answer: {language_level}, i.e. {language_level_context}
- The purpose of the conversation is to be {conversation_purpose}, i.e. {conversation_purpose_context}
- We expect you to answer with the following depth: {response_depth}, i.e. {response_depth_context}
{custom_description}
Use the following {language} in your communication.
If the question cannot be answered using the given context, answer "I have insufficient information to answer this
question." and give the appropriate indication.
Context:
$$$
{context}
$$$
History:
€€€
{history}
€€€
expected_output: >
Your answer.
metadata:

18
config/tasks/globals/SPIN_DETECT_TASK/1.0.0.yaml
View File

@@ -1,18 +0,0 @@
version: "1.0.0"
name: "SPIN Information Detection"
task_description: >
Complement the historic SPIN context (in between triple backquotes) with information found in the latest reply of the
end user.
{custom_description}
Use the following {tenant_language} to define the SPIN-elements.
Historic SPIN:
```{historic_spin}```
Latest reply:
{query}
expected_output: >
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task that performs SPIN Information Detection"
changes: "Initial version"

20
config/tasks/globals/SPIN_QUESTIONS_TASK/1.0.0.yaml
View File

@@ -1,20 +0,0 @@
version: "1.0.0"
name: "SPIN Question Identification"
task_description: >
Revise the final SPIN provided by your colleague, and ensure no information is lost from the histoic SPIN and the
latest reply from the user. Define the top questions that need to be asked to understand the full SPIN context
of the customer. If you think this user could be a potential customer, please indicate so.
{custom_description}
Use the following {tenant_language} to define the SPIN-elements. If you have a satisfying SPIN context, just skip and
don't ask for more information or confirmation.
Historic SPIN:
```{historic_spin}```
Latest reply:
{query}
expected_output: >
metadata:
author: "Josako"
date_added: "2025-01-08"
description: "A Task that identifies questions to complete the SPIN context in a conversation"
changes: "Initial version"