A strictly vetted directory of open-source artificial intelligence repositories, local LLM engines, multi-agent frameworks, and vector search tooling. Annotated with personal engineering takes from production builds.
Curated Open-Source AI Repositories
samber/cc-skills-golang ↗
Go-specific agentic skills for coding assistants: conventions, testing, security, observability, architecture, framework guides. Human-edited and tested with a claimed plus-41-point eval gain. When NOT to use: skip for single-language Python shops — loading Go conventions into context burns tokens with zero return outside Go repos.
“I keep one Go service (the telemetry collector fronting my ledger) beside a PHP and Python stack, and generic agents write unidiomatic Go without guardrails. The testing skill standardized table tests across that service in one sitting from Junagadh. I load these only inside the Go repo via directory-scoped config so Python work never pays the context tax.”
tetherto/qvac ↗
Local-first peer-to-peer AI: JS/TS plus Python clients running GGUF chat, RAG, speech, translation, and image tasks on Linux, macOS, Windows, Android, iOS, or delegating to peers; OpenAI-compatible HTTP server. When NOT to use: skip where sub-100ms cloud latency with frontier models is required — on-device quants trade raw quality for privacy and offline reach.
“I ship for Gujarat shops with patchy 4G where cloud calls die mid-quotation, so on-device GGUF plus peer delegation is a real deployment shape. I tested the OpenAI-compatible server as a drop-in for my Pi 5 harness at 62 tok/s and kept 80 percent of calls inside the VPC. I still route final drafting to cloud models — small quants draft, frontier models close.”
petergyang/no-ai-slop ↗
Filter stripping 20-plus AI writing patterns (throat-clearing openers, hype verbs, robotic transitions) from drafts. When NOT to use: skip as a sole quality gate — pattern removal polishes surface style but cannot add missing proof, metrics, or first-person testing.
“I run a 90-pattern blocklist as an automated audit gate before anything goes live from Junagadh, so a standalone slop remover slots into my pipeline as first pass. It flagged three throat-clearers in one founder-story draft last week that my own pass missed. Final sign-off stays human — no filter can verify a P95 number or a rupee price.”
bojieli/ai-agent-book ↗
AI Agent Book — full open book on agent design principles plus engineering practice: complete text, compiled PDF, per-chapter code. Python, +10.6k stars week Sep 18 2026 (24.7k total).
“I point Junagadh interns at ai-engineering-from-scratch before touching client work, and this is the missing second textbook — design principles plus engineering practice with code per chapter. Rebuilding planners and handoffs by hand removes the magic that causes production surprises. Two of my interns would read this before their first MCP tool.”
every-app/open-seo ↗
Open SEO — open-source alternative to Semrush and Ahrefs. TS, weekly trending +2.3-2.8k Sep 2026 (15.8k stars).
“I run AEO plus SEO for Gujarat SMEs where Rs 8k monthly SaaS seats eat the retainer. A self-hosted Semrush-shaped alternative I can point at deepakbagada.in citations plus AI Overview tracking is the tool my 90-day ledger wants. I would trial it against my PAA-mapping workflow for one client quarter before moving spend.”
tashfeenahmed/freellmapi ↗
FreeLLMAPI — 7.4 billion tokens per month across 34 free LLM providers. TS, weekly trending +3.6k Sep 2026 (23.3k stars).
“I route bulk batches to Sol plus GLM Flash plus Kimi on rupee math from Junagadh, and free-tier aggregation is the missing bottom rung for eval replays. Thirty-four providers behind one gateway means nightly promptfoo runs stop touching paid quotas. I would sandbox it for eval traffic only — never client money paths — with per-provider caps logged to the ledger.”
1jehuang/jcode ↗
JCode — the most RAM-efficient harness. Rust, 12.8k stars, +2.5k week Sep 18 2026.
“I test minimal harnesses on a Rs 6K VPS where every extra dependency breaks deploys, and RAM efficiency is the metric nobody benchmarks. JCode claiming the leanest harness in Rust is the exact shape I would run on Pi 5 edge agents at 62 tok/s beside GenericAgent. I would measure its context use against my 30K-context discipline before trusting the claim.”
CoreBunch/Instatic ↗
Instatic — open-source alternative to Webflow, Framer and WordPress. Agentic self-hosted visual CMS outputting clean static pages: users, roles, plugins, content, database. TS, +2.8k week (6.4k stars).
“I ship SME brochure sites from Junagadh at Rs 55-85K where clients then pay WordPress plugin ransom forever. An agentic visual CMS emitting clean static pages with users plus roles plus content built in is the handover shape my care plans want — 60ms TTFB cached, nothing to patch at 2am. I would trial it for brochure-only clients while Laravel keeps money logic.”
aipoch/open-science ↗
Open Science — open-source, local-first, model-agnostic AI research workbench for macOS, Windows, Linux. Scientific agents, Python/R notebooks, data connectors, reproducible provenance. TS (3.8k stars).
“I run RAG for client docs where plain vector search misses multi-hop answers, and provenance is the gap — which notebook cell plus which source produced this claim. Open-science wiring agents to Python/R notebooks with reproducible trails is the audit shape my 90-day ledger demands. I would trial it for Kimi-class research runs before trusting outputs in client reports.”
AprilNEA/OpenLogi ↗
OpenLogi — native, local-first alternative stack. Rust, 18.2k stars, +2.5k week Sep 18 2026.
“I prefer local-first for Gujarat client demos on patchy 4G, and my Pi 5 harness at 62 tok/s proves small plus local wins attention. OpenLogi packaging that philosophy natively in Rust is the install shape I would hand a Rajkot engineering client tired of cloud bills. I would benchmark it against my VPS parity rig before recommending the switch.”
ConardLi/garden-skills ↗
Garden Skills — open-source skills collection: web design, knowledge retrieval, image generation and more. 12.5k stars (verified via API Sep 19 2026).
“I maintain a 41-skill Junagadh catalog where each entry earns its place through client revenue. Garden-skills spanning design plus retrieval plus image generation is the cross-domain shelf I would audit skill by skill — keep the retrieval entries that beat my pgvector 42ms baseline, drop the rest. I install with tight approvals after the broad-permission audit, same gate as every skill.”
OpenWhispr/openwhispr ↗
OpenWhispr — voice-to-text dictation app with local Parakeet/Whisper plus cloud BYOK. Privacy-first, cross-platform. JS, 7.3k stars.
“I run client calls where DPDP forbids cloud recording, and meetily already covers meetings locally. OpenWhispr adding everyday dictation with local models plus BYOK fallback is the desk companion I would give founders who think faster than they type. I would keep it on-device only for clientnamed work and log consent the same way as voice notes.”
cursor/plugins ↗
Cursor plugins — official plugin specification plus plugins. TS, 6.3k stars, +1.4k week Sep 18 2026.
“I wire Cursor seats for Gujarat SME pilots where repeatable deploys beat hand-rolled prompts. The official plugin spec is what I would pin every Cursor setup to — read the spec, ship one verified plugin per seat, refuse the rest. I test new entries against my Rs 6K VPS harness before they touch client repos.”
likec4/likec4 ↗
LikeC4 — architecture diagrams as code. TS, 5.2k stars, +1.2k week Sep 18 2026.
“I publish architecture blueprints where Mermaid slop embarrasses the engineering. LikeC4 doing diagrams as reviewable code sits exactly between my hand-drawn SVGs and generated galleries — diffable, versioned, honest. I would trial it for the next three blueprint pages and keep whichever renders cleaner at mobile widths.”
asciimoo/hister ↗
Hister — your own search engine. Go, 3.4k stars, trending Sep 18 2026.
“I self-host SearXNG so client queries never leak, and hister as a personal engine in Go is the lighter sibling I would test for single-seat research. One binary, own index, zero third parties — the shape my DPDP-sensitive briefs want. I would run it beside SearXNG for a month and keep the one with better Gujarati recall.”
agegr/pi-web ↗
Pi-Web — web UI for the pi coding agent. TS, 3.1k stars, +1.4k week Sep 18 2026.
“I run pi-family harnesses where TUI-only keeps non-technical co-founders locked out. Pi-web putting the pi agent behind a browser UI is the delegation surface my Rajkot clients could actually tap — approve, steer, watch. I would trial it on staging with the same HITL queue as my terminal runs before any owner touches it.”