Curated Open-Source AI Repositories

“Gives autonomous agents unmetered, privacy-first web search capabilities without expensive SERP API quotas.”

“For IndiaAI mission labs & Gujarat universities — train small BharatGen-style models on ₹65/hr GPUs without closed APIs.”

“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.”

“Monitor n8n + pgvector + India GPU VPS in real time — alert on token spike before ₹15L bill hits (FinOps).”

“Most RAG is vector-only and forgets why an answer exists. Semantica adds graph-native provenance so an audit can trace context lineage — that is why it held #1 on Aug 11 before agency-agents reclaimed it. Rating: 4.3/5. When NOT to use: Skip for tiny single-doc Q&A — graph provenance pays off at team scale and compliance audits, not at “answer this PDF”.”

“Code search that actually understands imports, call graphs and cross-file edges — Tree-sitter + Memgraph beats chunk-embedding alone for refactoring. The +7 rank rebound #13→#6 on Aug 12 tracks real dev pain. Rating: 4.5/5. When NOT to use: Overkill for single-file scripts — gains show at multi-language monorepo scale where structure matters.”

“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.”

“For SME RAG I need provenance + eval — OpenKB bundles ingestion, vector store and retrieval scoring openly. #1 Trending AI Memory Aug 29 at 4.0k (+930 today) because teams are done with black-box RAG. Rating: 4.3/5. When NOT to use: Skip for single-PDF chat — OpenKB pays off as a shared org KB, not ad-hoc demo.”

“I ship Junagadh VPCs with 90-day ledger — ODS saves me 3 hours per Pi 5 setup by wiring Ollama, WebUI, and pgvector correctly on first try. Its compose template is what I fork for client on-prem installs.”

“I audit Laravel 13 codebases with pgvector — trace-mcp cuts my exploration from 42 min to one MCP call with typed traces. Its observation-driven approach pairs exactly with my Pydantic+OPA+HITL ledger.”

“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 ship offline-first Pi setups — Wand-Enhancer reminded me how much UX matters when wrapping a local binary. Its local config layer is the same idea I use for Pi 5 62 tok/s agent UX over 4G.”

“I ship both stacks with one ledger — pmo-platform is the first PMO governance skill that maps to my 90-day OTel JSONL. Its 13-stage pipe mirrors how I gate ₹15K+ Razorpay calls via OPA+HITL before irreversible.”

“I keep 90-day JSONL per tenant — ECC is the smallest repo that gets agent memory + eval right together. Its performance harness is what I benchmark my 500-sample replay against.”

“I run day-in-life 06:00-22:00 from Junagadh — this skill wires my local Super Productivity REST to Claude Code so my 18:00 OTel review auto-creates tasks. Its local-first API pattern is exactly my HITL card UX.”