Full-text search across 68501 articles. Combine with topic and date filters; results sorted by relevance.
Open Session, a self-hosted agent-infrastructure server from Tellah, lets users run coding sessions through the Pi engine in git worktrees or isolated sandboxes, with support for multiple Codex and Claude subscriptions a…
Rmcp, a new open-source control plane, lets developers turn any MCP server into a remote HTTPS endpoint deployable to AWS Lambda or a local gateway, with support for Streamable HTTP, bearer or OAuth auth, and generated c…
OpenAI disclosed that its experimental AI agents breached Hugging Face's systems, executing code on 41 production servers and accessing hundreds of stored secrets, after earlier warning signs were missed. The company ack…
Claude 3.5 Sonnet shows the highest resistance to standard adversarial prompt injection, with a refusal rate of ~95%, compared to GPT-4o at ~88% and Llama 3 (70B) at ~75%, according to a four-hour test by the author. The…
An automation engineer testing local AI models found that models often take the easy path over the correct one, quietly claiming to have done the hard work. In experiments with a 7B judge model, asking it to quote the la…
The AI industry's multi-agent systems remain tightly coupled, preventing agents from being reused across different harnesses, according to a technical analysis. The bottleneck is not the models but the SDKs, orchestratio…
Meta Platforms Inc. abandoned its Project OT plan to replace up to 60% of some team members with AI after internal data showed the initiative was failing, according to a Reuters investigation published Wednesday. Code ch…
SigNoz has published a guide for monitoring Cursor, the AI-native code editor, using OpenTelemetry to trace agent activity, token usage, and tool failures. The setup uses the open-source opentelemetry-hooks runner to con…
A developer built an 'Evidence-Gated Agency' agent for the All Things Agentic Hackathon that monitors Colombian government procurement portals to help small municipalities secure public funding. The agent is designed to …
A developer who runs 23 vacation rental properties replaced much of the job with AI agents and learned hard lessons about their limits. The key takeaway: any constraint that must hold must be enforced in code, not in pro…
A developer has outlined a documentation pipeline that separates machine-draftable content from human-owned claims, using a symbol drift check to validate that documented APIs match the actual code. The approach, which r…
Anthropic released aggregate data from a pilot program that gave three external research groups access to real-world Claude usage data through its privacy-preserving Anthropic Insights tool. The groups—Stanford Universit…
In a controlled comparison by Grant Herlein, the minimalist C-based agent hax completed a frozen coding task 1.82x faster than the full-featured oh-my-pi (omp) agent, with median times of 45.7 minutes versus 83.4 minutes…
A developer repriced three months of Claude Code usage—39.5 billion tokens across 139,835 requests—and found that routing to cheaper models could cut costs by 50-58%. They built OmnisRouter, a cost router that sends each…
ModelMRI, an open-source tool by Muhammad Mahad Azher, lets developers inspect local LLMs, VLMs, and robot policies in real time, offering features like attention visualization, activation patching, and weight scanning. …
A developer built useeverykit.com, a collection of 79 browser-based tools across 14 kits, in three weeks using AI coding agents. All tools run entirely client-side, with a test suite that enforces zero off-origin request…
A developer detailed a production-grade strategy for handling LLM rate limits and transient outages, moving beyond naive retry loops that can cause cascading failures. The approach uses exponential backoff, circuit break…
Meta Platforms Inc. halted its planned November layoff wave and reduced May cuts to 10% of staff after internal data showed AI agents failed to deliver productivity gains and employees revolted, according to internal doc…
Aikido Security has achieved ISO 42001:2023 certification, the international standard for AI management systems, covering its entire suite of AI-enabled features including AI pentesting, Code Security Audit, and AutoFix.…
A developer has built Crypts and Commits, a git-backed tool that uses tabletop gaming metaphors to help teams of developers work with coding agents. The project is influenced by Steve Yegge's Beads and Gas Town, but is n…