Skip to content
View inbharatai's full-sized avatar

Block or report inbharatai

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
inbharatai/README.md
InBharat AI

We build practical AI systems that turn language, context, and workflows into useful software.

Website KathaKitaab JAK Swarm JAK Shield Phoring

Followers Products and tools Focus


What We Build

InBharat AI is a product studio focused on agentic AI, AI safety, decision intelligence, and multilingual software for India-first use cases.

We care about systems that work outside a demo:

  • natural-language workflows that create useful output;
  • human approvals before high-risk actions;
  • audit trails and evidence, not invisible magic;
  • multilingual and voice-first experiences for Bharat;
  • open-source foundations where transparency matters.

We are intentionally careful with claims. Some projects are live, some are public beta, some have public repositories, and some are private product builds. The descriptions below reflect the current product posture.


Flagship Work

Interactive AI storybook for Indian epics and children’s stories.

KathaKitaab is a live 2.5D story experience where readers can tap characters and objects to branch stories in real time, with AI-generated scenes, narration, and music.

Core areas: interactive storytelling, Indian epics, children’s learning, real-time branching, multimodal generation.

Website

Closed-loop AI workflow operator for founder-led teams.

JAK turns plain-English tasks into planned, role-aware workflows with approval and audit surfaces. The current public posture is public beta: strong foundations, active hardening, and honest product-proof work in progress.

Core areas: multi-agent orchestration, role routing, workflow traces, approvals, builder surface, company-memory direction.

Website · Repository

Safety gateway for AI agents and tool calls.

JAK Shield is the separate security layer for policy checks, PII handling, prompt-injection defense, approval boundaries, and signed decisions before agents touch real tools.

Core areas: MCP-native safety, policy engine, prompt-injection scanning, sensitive-data controls, auditability.

Repository

Open-source decision intelligence.

Phoring helps teams move from documents to structured reasoning: ingest materials, build knowledge graphs, run scenario simulations, and produce source-cited forecasts.

Core areas: document intelligence, knowledge graphs, scenario planning, forecast reports, confidence tagging.

Website · Repository

AI social-media workflow engine.

SocialFlow focuses on planning, drafting, brand voice, image generation, and platform-aware social content workflows for solo builders and small teams.

Core areas: campaign workflow, content drafts, brand kits, platform adapters, local/open-source automation.

Repository

Voice-first AI advisor and product surface for Bharat.

InBharat AI is the home surface for Bharat-first AI workflows, consulting, and product experiments across language, research, education, productivity, and automation.

Core areas: multilingual UX, voice-first workflows, AI consulting, product prototyping, India-first access.

Website


Products and Tools

Project What it is Current posture
KathaKitaab 2.5D AI storybook where characters and objects can branch the story in real time Live product; private repo
JAK Swarm Closed-loop AI workflow operator for planned, approved, auditable agent work Public beta; public repo
JAK Shield Agent safety gateway for policy checks, PII controls, prompt-injection defense, and approval boundaries Public repo
Phoring Decision-intelligence system for documents, knowledge graphs, scenario simulations, and cited forecasts Public repo
SocialFlow Social-media workflow engine for planning, drafting, brand voice, image generation, and platform-aware automation Public repo
InBharat AI Voice-first AI advisor and product surface for Bharat-first workflows Live product; private repo
UniAssist.ai Education guidance surface for university discovery, applications, scholarships, and career planning Live product surface; public project page
TestsPrep.in AI test-prep surface for competitive exams, learning workflows, and student practice Live product surface; public project page
SwasthyaScore AI Mobile-first bilingual NCD screening concept for obesity, diabetes, BP, and risk scoring Private product build
RHCF Seva AI-powered field assistant for community health workers Private product build
Sahaayak Seva Public-health field-assistant concept for Anganwadi and community workflows Public concept repo
Sahaayak AI Multilingual personal AI OS concept for voice, chat, OCR, notes, news, and reminders Public concept repo
Agent Arcade Agent cockpit for observing, debugging, and controlling agent runs across frameworks Public prototype / reference implementation
OpenClaw AI CMO Local-first AI marketing workflow engine for solo builders Public open-source project
OpenClawFix Installer/configuration helper for OpenClaw setup workflows Private/internal tool
Claude Skills Skills collection for AI coding, local automation, research, and product-building workflows Public reference collection

Operating Principles

Principle What it means in practice
Truth over hype Do not present mock-only surfaces as production systems.
Human control High-risk agent actions should pause for approval.
Evidence by default Workflows should leave traces, logs, screenshots, or audit records.
Bharat-first UX Voice, multilingual, low-friction interfaces matter as much as model quality.
Safety as product PII, prompt injection, policy gates, and signed decisions are product features, not afterthoughts.
Open foundations Public repositories should be useful, inspectable, and honest about maturity.

Technology Areas

TypeScript Next.js Fastify Prisma Python Playwright OpenAI MCP

  • Agent orchestration and workflow runtime design.
  • AI safety gates and tool-call policy enforcement.
  • Browser automation and human-style product testing.
  • Document intelligence, graphs, and source-cited reasoning.
  • Multilingual, voice-first, India-first product UX.

Build With Us

We are especially interested in:

  • founders who want AI to execute real workflows safely;
  • teams building with agents, MCP, or approval-heavy automation;
  • public-interest and Bharat-scale use cases in health, education, productivity, and access;
  • open-source collaborators who care about truth, safety, and practical usefulness.

If the problem is real and the workflow matters, we want to build the working version.

Website · KathaKitaab · JAK Swarm · JAK Shield · Phoring · SocialFlow


Last reviewed for claim accuracy: 2026-05-22.

Pinned Loading

  1. agent-arcade-gateway agent-arcade-gateway Public

    Agent cockpit for observing, debugging, and controlling AI agent runs across frameworks, with traces, replay, cost views, and remote-control experiments.

    TypeScript 5