~/bookmarks

A Claude Code skill bundle for bug hunting and external red-team work - 82 skills, 15 slash commands, 681 disclosed-report patterns curated across 24 core vulnerability classes, plus enterprise ide…

A custom local engine that improves prefill and decode throughput

JetKVM native desktop GUI client. Contribute to lkarlslund/jetkvm-desktop development by creating an account on GitHub.

Behavioral differences and prompting patterns for Claude Fable 5.1 and Claude Mythos 5.1, covering effort, progress updates, tool-call batching, conversation history, writing style, formatting, task completion, compaction summaries, scope and test coverage, search triggering, safeguard false positives, file edits, long outputs, subagents, and vision.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Build machine learning pipelines visually with Tangle. Drag-and-drop editor for creating, managing, and executing ML workflows.

Modern systems in artificial intelligence (AI) somehow excel in domains for which they seem poorly suited. Intelligence has traditionally been modeled as operating over structured combinations of symbols, such as logical formulas. However, the strongest modern AI systems are based on neural networks, which instead represent information in continuous vectors. Vectors seem inadequate for capturing the structure of language, logic, and other cognitive domains, yet neural networks achieve impressive performance in these areas. How do they do it? In this work, we propose a potential answer: Despite appearances, perhaps the internal representations of neural networks implicitly realize symbolic structure. In support of this hypothesis, we show that the vector representations of a variety of neural networks can be closely approximated with symbolic structures: we can replace the network’s entire representation-generating process with a closed-form equation instantiating a symbolic structure, and the network’s behavior remains largely unchanged. This finding holds for both small-scale neural networks trained to manipulate lists as well as large language models (LLMs) operating in four domains that are central in symbolic traditions: arithmetic, logic, computer code, and language. Further, our symbolic approximation allows us to modify an LLM’s behavior in targeted ways via precise interventions on its internal representations, showing that the LLM’s behavior is reliant on the symbolic structures we have identified. This work provides a potential way to reconcile longstanding symbolic conceptions of intelligence with the vector-based nature of modern AI.

A farm of real iPhones, run from your Mac. Open-source iOS device automation with live control, a Postgres-backed scheduler, and TikTok workflows. Self-hosted, Apache-2.0. - Git-Agni/prod-FARM-IOS-…

Turbo mode is the fastest and most accurate web search API in the ultra-low price class.

Kanban board for herdr — cards are prompts dispatched to AI agents in visible panes - nelsonPires5/herdr-board

Curated directory of terminal-native AI coding agents and the harnesses that orchestrate them. Covers open-source tools (Pi, OpenCode, Aider, Goose), platform agents (Claude Code, Codex, Gemini CLI…

Apply to the Alpha Fellowship from a16z speedrun: capital before a team or an idea exists, for highly technical students and recent grads. 8 weeks, in person.

Learn more about the Sprites ecosystem: https://t.co/ycaws3u6qQ

Sentrux is a Rust-based architectural sensor that helps AI agents close the feedback loop and recursively improve code quality through real-time scanning. https://t.co/XzsY0rukvP

You don’t write AGENTS.md. You train it with gradient descent. - kunchenguid/backpass

Practice system design interviews with a free browser simulator. 10K+ RPS, 28 chaos scenarios, AWS cost estimates, 56 blueprints. No signup. Rust + WebAssembly.

Build a system, load it until it breaks, and watch why. A system design simulator for learning distributed systems. - xevrion/breakscale

A simple tool for coordinating several AI agents. Contribute to unclebob/swarm-forge development by creating an account on GitHub.

An AI skill that provides design intelligence for building professional UI/UX across multiple platforms. - nextlevelbuilder/ui-ux-pro-max-skill

Postgres Terminal. Contribute to pgrundev/pgterm development by creating an account on GitHub.

Postgres intelligence for ai agents & apps. Contribute to pgrundev/pgbot development by creating an account on GitHub.

A experimental fork of llama.cpp with LLM inference in C/C++ - quimmedes/cafe-llama.cpp

GEO-first SEO skill for Claude Code. Comprehensive AI search optimization for any website — citability scoring, AI crawler analysis, brand authority, schema markup, platform-specific optimization,…

Give your LLM a supercomputer. Thousands of GPUs on the only PaaS designed for agent-driven machine learning: nodes by the minute, batch jobs, sweeps, and RL sandboxes that fork in milliseconds.