Curated Open-Source AI Repositories

“I run OpenClaw-style assistants for Gujarat SME clients where agents need memory plus sandboxed tools without a Python dependency mess. OpenFang ships one 32MB Rust binary with cold start under 200ms from my Junagadh lab tests, and its WASM dual-metered sandbox is the pre-deploy gate I enforce before any agent touches Razorpay keys.”

“I test minimal harnesses on a Rs 6K VPS where every extra dependency breaks deploys. GenericAgent runs under 30K context where others burn 200K-1M, and its crystallize-each-task-into-skill loop matches how I grow my 41-skill Junagadh catalog after every client rollout.”

“I ship browser-use automation for Rajkot RFQ inboxes where agents stall on logins and Cloudflare. Agent S3 beating human OSWorld at 72.6 percent is the first computer-use result I trust for unattended runs, and I study its ACI grounding loop from Junagadh before wiring any GUI step into client pipelines.”

“I run LangGraph plus CrewAI-pattern swarms behind one OPA gate in Junagadh, and cross-harness coordination is still manual glue. Ruflo as a meta-harness with adaptive memory plus RAG that speaks Claude Code, Codex and Hermes natively is the consolidation layer my multi-harness setup is missing. I would trial it against my LangGraph checkpoint rig on catalog QA first.”

“Intuitive role-based mental model (Researcher, Writer, Auditor) that makes complex multi-agent swarms easy to conceptualize.”

“India developers run Gemini on ₹0 infra for CLI automation — pairs with n8n and Opencode for Junagadh builds.”

“Groundbreaking architecture for autonomous agent-to-agent code generation and execution loops.”

“Fastest discovery for Gujarat SMEs — pick skill, drop into Pi/OpenCode, ship n8n + voice + RAG without building from zero.”

“India budget teams code with Claude without USD billing — perfect Junagadh fresher path to ₹6–12L AI jobs.”

“The pioneering repository that ignited the entire autonomous agent revolution in early 2023.”

“Gujarat teams spec in Gujarati/Hindi, generate Laravel + n8n code — keeps SOP + DPDP ledger aligned.”

“Legal is the hardest agent eval — Harvey Labs gives you real legal tasks + grading harness instead of toy MMLU. At ~1.1k stars it is early but the only domain-specific agent skill benchmark I trust. Rating: 4.2/5. When NOT to use: Not for general coding evals — use SWE-bench; Harvey is narrowly tuned for legal reasoning and contract tasks.”

“I build auditable workflows for Rajkot foundries — this SWMM loop shows how to make a domain simulator (EPA SWMM) auditable with memory-informed retries. Its HITL card before irreversible runs is the pattern I steal for GST filings.”

“I build Gujarat SME automations where P95 and HITL before irreversible matter — AutoHedge is the fintech swarm that taught me to gate trades the same way I gate Razorpay ₹15K+ with OPA+HITL. Its swarm + risk layer is the finance analog of my 90-day ledger.”

“I audit MCP skills where 26% request broad permissions per SkillSpector — Masriyan 22-skill pack gives my Claude Code MITRE ATT&CK + NIST CSF 2.0 mapping and an authorization gate before exploit. Its GRC/CSOC playbooks are the checklist I run before any OPA+HITL deploy.”

“I run Pydantic + HITL for client calls where recall matters — Natively gives me a local-first meeting copilot that stays inside VPC for DPDP. Its Stealth + note pipeline is the in-VPC fallback I offer when clients reject cloud.”