76 lines
5.7 KiB
Markdown
76 lines
5.7 KiB
Markdown
## Daily Digest - February 20, 2026
|
|
|
|
### 🤖 iOS AI Development
|
|
|
|
**Apple Foundation Models Framework Now Available**
|
|
Apple has released the Foundation Models framework giving developers direct access to the on-device foundation model at the core of Apple Intelligence. With native Swift support, you can tap into the model with as few as three lines of code to power features like text extraction, summarization, and more - all working without internet connectivity.
|
|
[Read more →](https://developer.apple.com/machine-learning/)
|
|
|
|
**SpeechAnalyzer Brings Advanced On-Device Transcription to iOS**
|
|
The all-new SpeechAnalyzer framework enables advanced, on-device transcription capabilities for your apps. Take advantage of speech recognition and saliency features for a variety of languages without sending audio data to the cloud.
|
|
[Read more →](https://developer.apple.com/machine-learning/)
|
|
|
|
**Core ML Updates for Vision and Document Recognition**
|
|
New updates to Core ML and Vision frameworks bring full-document text recognition and camera smudge detection to elevate your app's image analysis capabilities on Apple devices.
|
|
[Read more →](https://developer.apple.com/machine-learning/)
|
|
|
|
### 💻 AI Coding Assistants
|
|
|
|
**Cursor Launches Plugin Marketplace with Partners Including Figma, Stripe, AWS**
|
|
Cursor has introduced plugins that package skills, subagents, MCP servers, hooks, and rules into single installs. Initial partners include Amplitude, AWS, Figma, Linear, and Stripe, covering workflows across design, databases, payments, analytics, and deployment.
|
|
[Read more →](https://cursor.com/blog/marketplace)
|
|
|
|
**Cursor CLI Gets Cloud Handoff and ASCII Mermaid Diagrams**
|
|
The latest Cursor CLI release introduces the ability to hand off plans from CLI to cloud, inline rendering of ASCII diagrams from Mermaid code blocks, and improved keyboard shortcuts for plan navigation.
|
|
[Read more →](https://cursor.com/changelog)
|
|
|
|
**Stripe Releases Minions - One-Shot End-to-End Coding Agents**
|
|
Stripe has published Part 2 of their coding agents series, detailing "Minions" - their one-shot end-to-end coding agents that help automate development workflows.
|
|
[Read more →](https://stripe.dev/blog/minions-stripes-one-shot-end-to-end-codi)
|
|
|
|
**GitHub Copilot Now Supports Multiple LLMs and Custom Agents**
|
|
GitHub Copilot now lets developers choose from leading LLMs optimized for speed, accuracy, or cost. The platform supports custom agents and third-party MCP servers to extend functionality.
|
|
[Read more →](https://github.com/features/copilot)
|
|
|
|
### 🧠 Latest Coding Models
|
|
|
|
**Claude Opus 4.6 Released with Major Coding Improvements**
|
|
Anthropic has upgraded their smartest model - Opus 4.6 is now an industry-leading model for agentic coding, computer use, tool use, search, and finance, often winning by wide margins in benchmarks.
|
|
[Read more →](https://www.anthropic.com/news)
|
|
|
|
**Gemini 3.1 Pro Rolls Out with Advanced Reasoning**
|
|
Google's new Gemini 3.1 Pro AI model "represents a step forward in core reasoning" according to Google. The model is designed for tasks where a simple answer isn't enough, rolling out now in the Gemini app and NotebookLM.
|
|
[Read more →](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/)
|
|
|
|
**GGML.ai Joins Hugging Face to Advance Local AI**
|
|
GGML.ai, the organization behind llama.cpp, is joining Hugging Face to ensure the long-term progress of Local AI. This partnership strengthens the ecosystem for running AI models locally on consumer hardware.
|
|
[Read more →](https://github.com/ggml-org/llama.cpp/discussions/19759)
|
|
|
|
**Taalas Demonstrates Path to 17k tokens/sec Ubiquitous AI**
|
|
Taalas has shared research on achieving ubiquitous AI with breakthrough performance of 17,000 tokens per second, showing a potential path to making AI inference dramatically faster and more accessible.
|
|
[Read more →](https://taalas.com/the-path-to-ubiquitous-ai/)
|
|
|
|
### 🦾 OpenClaw Updates
|
|
|
|
**OpenClaw Mentioned in Major Security Report on Cline Vulnerability**
|
|
A hacker reportedly tricked Cline's Claude-powered workflow into installing OpenClaw on computers, highlighting the importance of verifying AI agent actions. The incident was covered by The Verge as part of broader AI security concerns.
|
|
[Read more →](https://www.theverge.com/ai-artificial-intelligence)
|
|
|
|
### 🚀 Digital Entrepreneurship
|
|
|
|
**Bootstrapping a $20k/mo AI Portfolio After VC-Backed Failure**
|
|
An inspiring story of an entrepreneur who built a $20,000/month AI portfolio through bootstrapping after their VC-backed company failed. The approach focuses on sustainable revenue over growth-at-all-costs.
|
|
[Read more →](https://www.indiehackers.com/post/tech/bootstrapping-a-20k-mo-ai-portfolio-after-his-vc-backed-company-failed-rQxwZBD9xWVgfHhIxvbJ)
|
|
|
|
**Hitting $10k/mo by Using Agency as Testing Ground and Distribution**
|
|
A developer shares how they reached $10,000/month by using their agency as both a testing ground for product ideas and a distribution channel for their SaaS products.
|
|
[Read more →](https://www.indiehackers.com/post/tech/hitting-10k-mo-by-using-an-agency-as-both-testing-ground-and-distribution-FF8kooe4FWGH9sHjVrT3)
|
|
|
|
**Bazzly: Your SaaS Needs a Distribution Habit, Not Just Strategy**
|
|
A new product launching with the insight that SaaS companies don't need complex marketing strategies - they need consistent distribution habits to reach customers effectively.
|
|
[Read more →](https://www.indiehackers.com/product/bazzly)
|
|
|
|
**LLM Eagle: New LLM Visibility Tool for Developers**
|
|
A new indie hacking project creating an LLM visibility tool that focuses on simplicity and developer experience, aiming to solve monitoring challenges without the complexity of enterprise solutions.
|
|
[Read more →](https://www.indiehackers.com/product/llm-eagle)
|