Skip to content

Add Ekatra Learning — AI-powered micro-learning company (6 agents, 10 skills)#5

Open
iabheejit wants to merge 2 commits intopaperclipai:mainfrom
iabheejit:add-ekatra-learning
Open

Add Ekatra Learning — AI-powered micro-learning company (6 agents, 10 skills)#5
iabheejit wants to merge 2 commits intopaperclipai:mainfrom
iabheejit:add-ekatra-learning

Conversation

@iabheejit
Copy link
Copy Markdown

What is Ekatra Learning?

An AI-powered micro-learning company template that delivers hyper-personalized courses through WhatsApp, offline Raspberry Pi hubs, and web — achieving 65% completion rates versus the industry standard 5-15%.

Built from Ekatra, an operational L&D platform targeting 1 billion underserved learners globally (refugee camps, rural India, FMCG field sales teams).

Agents (6)

Agent Role
CEO Strategy, client acquisition, product direction
Content Designer Generates 3-day micro-courses from raw material (PDFs, transcripts, SME notes)
Learning Delivery Agent WhatsApp course deployment via Twilio, offline Pi provisioning, spaced repetition
Analytics Agent Completion tracking, BKT knowledge tracing, enterprise ROI reports
Outreach Agent Prospect research, qualification, outreach sequences
Platform Engineer Cloud infra, offline device fleet, deployment pipelines

Skills (10)

Skill Description
generate-microcourse Raw content to 3-day WhatsApp course JSON with quizzes and flashcards
whatsapp-deploy Enrollment, daily scheduling, progress tracking, re-engagement via Twilio
spaced-repetition SM-2/FSRS flashcard scheduling for long-term retention
learner-analytics Completion aggregation, BKT mastery estimation, cohort analysis
content-audit Quality review for accuracy, reading level, cultural sensitivity, WhatsApp formatting
prospect-research L&D client identification and qualification (FMCG, NGO, enterprise)
offline-deploy Raspberry Pi provisioning with TinyLlama, WiFi hotspot, sync
course-localize Language/cultural adaptation, literacy-level adjustment, audio scripts
engagement-report Weekly/monthly client reports with completion funnels and ROI calculations
client-onboard Enterprise onboarding: needs assessment, pilot, scale-up

Why this is useful

This is the first L&D / education-focused company template in the collection. It packages a proven operational model (65% completion rates, $0.40-0.80/learner/month unit economics) into a reusable template that anyone can spin up to run an AI-powered training company — whether for corporate workforce development, NGO education programs, or government skilling initiatives.

Structure

Follows the agentcompanies/v1 spec:

  • COMPANY.md — metadata, description, goals, workflow narrative
  • agents/ — 6 agents with AGENTS.md files (role, triggers, handoffs, principles)
  • skills/ — 10 skills with SKILL.md files (detailed workflows, inputs, outputs, quality checks)
  • .paperclip.yaml — runtime config (Twilio, AWS, Azure secrets)
  • LICENSE — MIT
  • README.md — overview, org chart, tables, import command

… skills)

An L&D company template that delivers hyper-personalized courses through
WhatsApp, offline Raspberry Pi hubs, and web — achieving 65% completion
rates versus the industry standard 5-15%.

Agents: CEO, Content Designer, Learning Delivery, Analytics, Outreach,
Platform Engineer

Skills: generate-microcourse, whatsapp-deploy, spaced-repetition,
learner-analytics, content-audit, prospect-research, offline-deploy,
course-localize, engagement-report, client-onboard

Built from https://github.com/nicholasguo/ekatra
- Rename package slug to socrates-learning-bot
- Replace 6-agent L&D platform with 4-agent chatbot company
  (CEO, Marketing, Learning, Technology)
- Replace 10 skills with 5 focused skills
  (generate-lesson, send-message, send-quiz, prospect-outreach, deploy-bot)
- Channels: WhatsApp (Twilio) + Telegram Bot API
- Align COMPANY.md, .paperclip.yaml, and README.md with new design
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant