Prototype Chrome extension (MV3) for high-performance network interception across tabs and frames using SharedWorker, hidden iframes, and provider-based DLP-style detection. - vsromanc/framelink

The native iOS companion for Plausible Analytics. Free to download.

⚒️ The open source dark software factory for expert engineers. - fabro-sh/fabro

SSH login with SSO. Lightweight alternative to Teleport & Smallstep - azophy/sshifu

[March 2013] On March 2, 2013, VP launched a billboard campaign throughout Washington DC with the US Campaign to End the Israeli Occupation. As the American Israel Public Affairs Committee (AIPAC) met for its annual Congressional lobby, the campaign called for an end to the $30billion in US military aid promised to Israel over the coming decade. This is one of three posters highlighting the ways in which US funding to Israel could instead be used to fund communities' needs. Posters were seen by commuters on the Washington DC Metro and on a billboard affixed to a truck following AIPAC conference attendees.

yoink is an AI agent that removes complex dependencies by reimplementing only what you need. - theogbrand/yoink

Agentic applications based on large language models increasingly rely on multi-step interaction loops involving planning, action execution, and environment feedback. While such systems are now deployed at scale, improving them post-deployment remains challenging. Agent trajectories are voluminous and non-deterministic, and reviewing each one, whether through human review or auxiliary LLMs, is slow and cost-prohibitive. We propose a lightweight, signal-based framework for triaging agentic interaction trajectories. Our approach computes cheap, broadly applicable signals from live interactions and attaches them as structured attributes for trajectory triage, identifying interactions likely to be informative without affecting online agent behavior. We organize signals into a coarse-grained taxonomy spanning interaction (misalignment, stagnation, disengagement, satisfaction), execution (failure, loop), and environment (exhaustion), designed for computation without model calls. In a controlled annotation study on $τ$-bench, a widely used benchmark for tool-augmented agent evaluation, we show that signal-based sampling achieves an 82\% informativeness rate compared to 74\% for heuristic filtering and 54\% for random sampling, with a 1.52x efficiency gain per informative trajectory. The advantage is robust across reward strata and task domains, confirming that signals provide genuine per-trajectory informativeness gains rather than merely oversampling obvious failures. These results show that lightweight signals can serve as practical sampling infrastructure for agentic systems, and suggest a path toward preference data construction and post-deployment optimization.

The social network for AI agents. Post, follow, react, and discover — via TypeScript SDK, CLI, MCP Server, or REST API.

Kasava is the AI-native platform purpose-built for product development. Plan, build, and ship with AI-powered chat, call analysis, and plan generation.

Lifecycle-first vanilla JS framework. Explicit ownership, deterministic teardown, no build step. - BleedingXiko/RAGOT

For small businesses, where AI agents post, reconcile, and organize real bookkeeping work.

SeekLink — hybrid semantic search and link discovery MCP server for Obsidian vaults. Fully local, native CJK support. - simonsysun/seeklink

A self-hosted dashboard and organizer for Excalidraw with multi-user collaboration and scoped sharing. - ZimengXiong/ExcaliDash

The first native macOS SDK for NXP NTAG 424 DNA NFC authentication - jetnoir/ntag424-macos