The Vaib Report · 2026-08-28
Daily Digest – 2026-08-28
Daily curated repos and tools for solo founders, AI/B2B SaaS builders, and small business tools. 1217 repos evaluated, 30 made the cut. Plus 39 picks from HuggingFace, Replicate, Papers with Code, npm, PyPI, GitLab, Ollama, and launch platforms.
Founder & Solopreneur Tools
sickn33/agentic-awesome-skills
Local agent control plane with 2000+ agentic skills, catalog discovery, and stack validation via CLI and MCP. (Python, 45.6k★)
★ 45,600
agno-agi/agno
Framework for building, running, and managing agent platforms with multi-provider LLM support. (Python, 42k★)
★ 41,957
mindsdb/mindshub
Unified workspace where open-source models execute tasks, swap models while keeping your work, Docker-ready setup. (Makefile, 39.7k★)
★ 39,656
wshobson/agents
Multi-harness agentic plugin marketplace for Claude Code, Codex, Cursor, GitHub Copilot with 93 plugins, 202 agents, one Markdown source. (Python, 39.2k★)
★ 39,228
Tool Sprawl & Alternatives
supertokens/supertokens-core
Open source Auth0 alternative, self-hosted authentication with email, OAuth, session management, Java-based. (Java, 15.3k★)
★ 15,288
yusufkaraaslan/Skill_Seekers
Convert docs, repos, and PDFs into Claude skills with automatic conflict detection and MCP integration. (Python, 14.8k★)
★ 14,848
sml2h3/ddddocr
General-purpose CAPTCHA and OCR recognition library, Python package. (Python, 14.7k★)
★ 14,681
tisfeng/Easydict
Native macOS dictionary and translator app supporting offline OCR, 10+ backends (Apple, OpenAI, DeepL, Google, Bing, Baidu). (Swift, 14.4k★)
★ 14,390
UGC, Social & Creative Marketing
Vincentwei1021/video-shotcraft
AI video skill for Claude Code: cinematic product videos with Remotion, 152 shot recipes, 209 motion previews. (TypeScript, 6.6k★)
★ 6,597
vllm-project/vllm-omni
Fast, cheap omni-modality model serving for text, image, audio, video, and diffusion inference with batching. (Python, 6.4k★)
★ 6,435
showlab/Awesome-Video-Diffusion
Curated list of video diffusion models for generation, editing, and restoration, no runnable code. (Multi, 5.8k★)
★ 5,762
salesforce/BLIP
Deprecated vision-language model for image captioning, VQA, and retrieval, archived in favor of LAVIS. (Jupyter Notebook, 5.7k★)
★ 5,716
apify/apify-mcp-server
MCP server giving agents access to thousands of Apify scrapers for data extraction from any website. (TypeScript, 5.3k★)
★ 5,275
Oddities
traefik/traefik
Cloud-native HTTP reverse proxy and load balancer for Docker, Kubernetes, and microservices. (Go, 64.6k★)
★ 64,636
helix-editor/helix
Modal Rust text editor inspired by Kakoune and Neovim, batteries-included with LSP and tree-sitter. (Rust, 46k★)
★ 45,971
tailscale/tailscale
WireGuard-based private networks with zero-config peer discovery, 2FA, cross-platform daemon. (Go, 35.7k★)
★ 35,708
freestylefly/awesome-gpt-image-2
Prompt-as-code library with 530+ image generation examples, 20+ industrial templates, reusable skills. (JavaScript, 24.2k★ ⚡1249★/day)
★ 24,219
every-app/open-seo
Open source Semrush alternative with keyword research, rank tracking, backlinks, site audit via MCP and AI skills. (TypeScript, 14.1k★ ⚡282★/day)
★ 14,136
Own Your Iron
xyproto/algernon
Single-binary web server with HTTP/2, QUIC, Lua, Teal, TypeScript, React, SQLite, PostgreSQL, live reload. (JavaScript, 3k★)
★ 3,027
thu-pacman/chitu
High-performance LLM serving framework for DeepSeek, Qwen, GLM, single card to multi-node clusters. (Python, 3k★)
★ 2,998
Dicklesworthstone/llm_aided_ocr
Enhance Tesseract OCR with local or API LLMs for error correction, smart chunking, markdown output. (Python, 3k★)
★ 2,993
MiniMax-AI/Mini-Agent
Reference agent implementation with M2.5 model, interleaved thinking, session memory, 15 Claude skills, MCP support. (Python, 3k★)
★ 2,988
vercel-labs/opensrc
Fetch and cache npm, PyPI, crates.io source code locally for AI agents to analyze dependency code. (Rust, 3k★)
★ 2,970
AI Cost & Token Economics
jia-gao/leanctx
Drop-in prompt compression cutting token bills 10-40% without code changes, runs locally, MIT licensed. (Python, 316★)
★ 316
project-koku/koku
Open source cloud cost management for AWS, Azure, GCP with insights, optimization recommendations, multicloud. (Python, 312★)
★ 312
kosty-cloud/kosty
Scan 30+ AWS services for cost waste, security gaps, and attack surface mapping, includes Bedrock audit. (Python, 271★)
★ 271
siara-cc/Unishox2
Hybrid compression for short Unicode strings, ultra-low memory footprint, Arduino-compatible, chat message compression. (C, 249★)
★ 249
HduSy/tokenscope
macOS menu-bar dashboard showing Claude CLI daily token usage, estimated cost, per-model/MCP/skill breakdown. (Rust, 246★)
★ 246
SeleniumHQ/selenium
Browser automation framework and W3C WebDriver implementation for cross-platform testing and scripting. (Java, 34.4k★)
★ 34,414
OpenAPITools/openapi-generator
Generate SDK, server stubs, and docs from OpenAPI specs, supports 70+ languages and frameworks. (Java, 26.7k★)
★ 26,695
Replicate Models
prunaai/p-video
Fast video generation with built-in draft mode for rapid creative iteration. Text-to-video, image-to-video, and audio-to-video in a single endpoint. (3.2M runs)
prunaai/z-image-turbo-lora
Lora version of Z-Image Turbo, which is a super fast text-to-image model of 6B parameters developed by Tongyi-MAI. (652K runs)
xai/grok-imagine-image-2
xAI's Grok Imagine Image 2.0 — text-to-image generation and editing with a quality control and output up to 2k (47K runs)
ultralytics/yolov8s-worldv2
Open-vocabulary YOLOv8s-WorldV2 detection — detect any classes you name. (17K runs)
nelsonjchen/op-replay-clipper
GPU accelerated replay renderer / video data clipper for comma.ai connect's openpilot route data. SEE README. (6K runs)
Ollama Models
llama3-gradient
This model extends LLama-3 8B's context length from 8k to over 1m tokens. (994.4K pulls)
meditron
Open-source medical large language model adapted from Llama 2 to the medical domain. (844.2K pulls)
nomic-embed-text-v2-moe
nomic-embed-text-v2-moe is a multilingual MoE text embedding model that excels at multilingual retrieval. (808K pulls)
command-r-plus
Command R+ is a powerful, scalable large language model purpose-built to excel at real-world enterprise use cases. (794.9K pulls)
mistral-small3.1
Building upon Mistral Small 3, Mistral Small 3.1 (2503) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance. (782.4K pulls)
npm Packages
pac-proxy-agent
A PAC file proxy `http.Agent` implementation for HTTP (29.2M/wk)
puppeteer-core
A high-level API to control headless Chrome over the DevTools Protocol (20.9M/wk)
puppeteer
A high-level API to control headless Chrome over the DevTools Protocol (12.1M/wk)
global-agent
Global HTTP/HTTPS proxy configurable using environment variables. (11.6M/wk)
cypress
Cypress is a next generation front end testing tool built for the modern web (7.6M/wk)
PyPI Packages
firecrawl
Python SDK for the Firecrawl API: web scraping, crawling, web search, and scientific literature search over a research paper index of PubMed, bioRxiv, medRxiv and arXiv abstracts (310K recent downloads)
dspy-ai
DSPy (136K recent downloads)
autogen
A programming framework for agentic AI (29K recent downloads)
HuggingFace Spaces
HuggingFace Models
argmaxinc/whisperkit-coreml
whisperkit-coreml (automatic-speech-recognition, 10.9M downloads)
timm/efficientnet_b3.ra2_in1k
efficientnet_b3.ra2_in1k (image-classification, 10.4M downloads)
Qwen/Qwen2.5-7B-Instruct
Qwen2.5-7B-Instruct (text-generation, 10.4M downloads)
jonatasgrosman/wav2vec2-large-xlsr-53-japanese
wav2vec2-large-xlsr-53-japanese (automatic-speech-recognition, 10.2M downloads)
pyannote/speaker-diarization-3.1
speaker-diarization-3.1 (automatic-speech-recognition, 9.4M downloads)
HuggingFace Datasets
New Launches
PageIndex
Accurate, trustworthy answers across professional documents (312 upvotes) [producthunt]
Caddi
Agent that builds agents by only showing your work only once (242 upvotes) [producthunt]
Microduck
A tiny open-source biped you can train yourself (224 upvotes) [producthunt]
OpenTIE and OpenXWA, Modern Ports of Tie Fighter and X-Wing Alliance
Show HN: OpenTIE and OpenXWA, Modern Ports of Tie Fighter and X-Wing Alliance (216 upvotes) [hackernews_showhn]
We built open OpenRouter that turns usage into a better model
Show HN: We built open OpenRouter that turns usage into a better model (207 upvotes) [hackernews_showhn]
Gemini Omni 1.1 Flash
Our newest multimodal model for video generation and editing (176 upvotes) [producthunt]
Voronoi Go
Show HN: Voronoi Go (153 upvotes) [hackernews_showhn]
Firecrawl Developer Index
A curated index of 70M+ artifacts for coding agents. (142 upvotes) [producthunt]