<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>Forem: David Nduonofit</title>
    <description>The latest articles on Forem by David Nduonofit (@david_nduonofit_4b96bc483).</description>
    <link>https://forem.com/david_nduonofit_4b96bc483</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3593501%2Fc781570d-7d68-4803-af82-eae453001730.png</url>
      <title>Forem: David Nduonofit</title>
      <link>https://forem.com/david_nduonofit_4b96bc483</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://forem.com/feed/david_nduonofit_4b96bc483"/>
    <language>en</language>
    <item>
      <title>HNG Stage 3 Task</title>
      <dc:creator>David Nduonofit</dc:creator>
      <pubDate>Mon, 03 Nov 2025 14:45:10 +0000</pubDate>
      <link>https://forem.com/david_nduonofit_4b96bc483/hng-stage-3-task-1fj4</link>
      <guid>https://forem.com/david_nduonofit_4b96bc483/hng-stage-3-task-1fj4</guid>
      <description>&lt;h1&gt;
  
  
  Building an AI Agent that conforms to A2A Standards
&lt;/h1&gt;

&lt;p&gt;During &lt;strong&gt;Stage 3 of the HNG Internship&lt;/strong&gt;, I built something that combined my faith and my love for code — an &lt;strong&gt;Agent-to-Agent (A2A) Christian Therapy API&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;This API, powered by &lt;strong&gt;FastAPI&lt;/strong&gt; and the &lt;strong&gt;JSON-RPC 2.0&lt;/strong&gt; standard, provides &lt;strong&gt;biblical therapy and encouragement&lt;/strong&gt; through the &lt;strong&gt;Word of God&lt;/strong&gt;. It’s designed to communicate with other intelligent agents on platforms like &lt;a href="https://telex.im" rel="noopener noreferrer"&gt;Telex&lt;/a&gt;, returning structured, faith-based responses to emotional or spiritual questions.&lt;/p&gt;




&lt;h2&gt;
  
  
  The Vision
&lt;/h2&gt;

&lt;p&gt;I wanted to see if an AI could provide &lt;strong&gt;comfort through scripture&lt;/strong&gt;, not just conversation. Instead of offering generic motivational replies, this agent responds with &lt;strong&gt;Bible verses, reflection, and hope&lt;/strong&gt; — much like a digital counsellor guided by faith.&lt;/p&gt;

&lt;p&gt;Here’s the response format from the &lt;code&gt;/a2a/therapist&lt;/code&gt; endpoint:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight json"&gt;&lt;code&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="w"&gt;
  &lt;/span&gt;&lt;span class="nl"&gt;"jsonrpc"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"2.0"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;&lt;span class="w"&gt;
  &lt;/span&gt;&lt;span class="nl"&gt;"id"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"request-001"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;&lt;span class="w"&gt;
  &lt;/span&gt;&lt;span class="nl"&gt;"result"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="w"&gt;
    &lt;/span&gt;&lt;span class="nl"&gt;"status"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="w"&gt;
      &lt;/span&gt;&lt;span class="nl"&gt;"state"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"completed"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;&lt;span class="w"&gt;
      &lt;/span&gt;&lt;span class="nl"&gt;"message"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="w"&gt;
        &lt;/span&gt;&lt;span class="nl"&gt;"role"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"agent"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;&lt;span class="w"&gt;
        &lt;/span&gt;&lt;span class="nl"&gt;"parts"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="w"&gt;
          &lt;/span&gt;&lt;span class="p"&gt;{&lt;/span&gt;&lt;span class="w"&gt;
            &lt;/span&gt;&lt;span class="nl"&gt;"kind"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;&lt;span class="w"&gt;
            &lt;/span&gt;&lt;span class="nl"&gt;"text"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;&lt;span class="w"&gt; &lt;/span&gt;&lt;span class="s2"&gt;"A detailed response stuff..."&lt;/span&gt;&lt;span class="w"&gt;
          &lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="w"&gt;
        &lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="w"&gt;
      &lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="w"&gt;
    &lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="w"&gt;
  &lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="w"&gt;
&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Each message is structured for &lt;strong&gt;A2A interoperability&lt;/strong&gt;, meaning other agents can easily parse and understand the response — just like humans understand emotional tone.&lt;/p&gt;




&lt;h2&gt;
  
  
  The Journey
&lt;/h2&gt;

&lt;p&gt;This project wasn’t without challenges.&lt;br&gt;
I ran into several issues — from malformed JSON-RPC structures to strict Pydantic model validation errors. Ensuring that every response matched the A2A standard took patience, but it taught me how &lt;strong&gt;AI agents communicate cleanly and predictably&lt;/strong&gt; across systems.&lt;/p&gt;




&lt;h2&gt;
  
  
  Why It Matters
&lt;/h2&gt;

&lt;p&gt;In a time when people turn to chatbots for emotional support, I wanted to create one that &lt;strong&gt;points back to God’s Word&lt;/strong&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  It’s not meant to replace real counseling, but to be a &lt;strong&gt;gentle companion&lt;/strong&gt; — offering reminders of peace, comfort, and faith in difficult times.
&lt;/h2&gt;

&lt;h2&gt;
  
  
  The Future
&lt;/h2&gt;

&lt;p&gt;I plan to enhance it with:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Emotional tone detection&lt;/li&gt;
&lt;li&gt;Multi-turn conversation memory&lt;/li&gt;
&lt;li&gt;Verse retrieval from different Bible translations&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;You can check out the full source code and setup details here:&lt;br&gt;
👉 &lt;a href="https://github.com/Forsaken324/ndus-hng-stage-three" rel="noopener noreferrer"&gt;GitHub – Forsaken324/ndus-hng-stage-three&lt;/a&gt;&lt;/p&gt;




</description>
      <category>learning</category>
    </item>
  </channel>
</rss>
