A curated collection of noteworthy tools, articles, and resources from across the web — searchable, organized, and enriched with AI analysis.
Showing 1-24 of 1252 bookmarks
All Stories
Local-first search across your workspace, built for humans and AI agents. - zvec-ai/zvec-grep
The Context Layer for unstructured data: typed, versioned datasets over S3, GCS, Azure - datachain-ai/datachain
Source content is unavailable because the publisher returned an access challenge.
LLM infradebugging and diagnostic tool. Contribute to day50-dev/llcat development by creating an account on GitHub.
Contribute to NVIDIA/flare-local-inference development by creating an account on GitHub.
A coding-agent skill for multi-phase security audits with independently verified, machine-readable findings - cloudflare/security-audit-skill
The open-source managed agent platform. A self-hostable alternative to Claude Managed Agents and OpenAI’s Agents API. - omnara-ai/omnara
The fastest way to run Qwen3.8-Flash-Next on Strix Halo (gfx1151) - peonist-ai/halogen-flash-server
Cheerio-compatible HTML/XML processing powered by Rust, with configurable parallel extraction. Built by context.dev. - context-dot-dev/rusteerio
Example of a durable Pi harness running on Cloudflare - earendil-works/pi-durable-cloudflare
Mirror is Reflection’s coding harness. Contribute to reflection-oss/mirror development by creating an account on GitHub.
One UI for model state, request history, API keys, routing rules, and proxy metrics — fronting llama-swap and any OpenAI- or Anthropic-compatible upstream.
An agentic coding launcher for Aperture. Contribute to tailscale/aperture-cli development by creating an account on GitHub.
Local-first session search, analytics, insights, and token use statistics for coding agents, supporting Claude Code, Codex, and more than 20 other agents. - kenn-io/agentsview
Contribute to togethercomputer/HybridAgent development by creating an account on GitHub.
How can we build an agent that thinks beyond human language, yet still acts through it in the external environment? Modern LLM agents still reason and interact through text, while latent reasoning offers an efficient internal medium but lacks native tool access and reasoning transparency. We bridge these two directions with Hybrid Reasoning Agents (HybridAgents), which dynamically navigate between latent reasoning and text-based interaction throughout a unified generation process. Architecturally, a lightweight HybridLink enables seamless transitions between text and latent modes, while the agent’s continuous working memory preserves lossless hybrid context across multiple turns. These designs preserve tool-integrated agentic capabilities while making latent thoughts directly inspectable. Building on this architecture, we further develop dedicated post-training algorithms for hybrid reasoning: HybridSFT stabilizes tool-integrated hybrid generation by teaching the backbone to coordinate with HybridLink, then HybridRL scales reasoning by learning when to think in latent space and when to act through text. We evaluate HybridAgent at 3 model scales across 11 benchmarks covering mathematics, science, medicine, search, and agentic reasoning. Compared with frontier LLM Agents and agentic frameworks spanning both text-based and latent methods, HybridAgent consistently improves performance by 10.4%, while delivering 35.5%-54.7% token reduction, 1.8x-2.6x faster inference, and 2.7x-3.7x cost savings. Code and data are available at https://github.com/togethercomputer/HybridAgent.
Standalone SearXNG CLI for coding agents. Contribute to nikvdp/searxng-ai-kit development by creating an account on GitHub.
Copy-on-write worktrees with memory that survives them - lukeed/lane
Outsource your thinking, not your understanding. A comprehension checkpoint for agent-assisted development. - parthh01/sphinx
A Git hook for agent-assisted development. Before your agent’s code ships, Sphinx asks you a question about what it actually does.
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of…
Feather is an agent that works inside the folders you already use. Ask for the model, the memo, or the deck, watch it work, and review every file it touched.
Agentic AI workspaces for the Enterprise.
I work on Claude. It’s an interesting experience: Every day I learn about yet another feat accomplished with or by Claude that I hadn’t considered, didn’t think possible, or simply lacked the curiosity and time to try. This month, it may have helped discover a new planet,
