Full-text search across 59534 articles. Combine with topic and date filters; results sorted by relevance.
Developer zhaoxuya520 has released reverse-skill, an open-source framework that routes AI coding agents through verified execution paths for cybersecurity tasks such as reverse engineering and vulnerability detection. Th…
Researchers propose STEP-KTODER, a framework for code preference optimization that defines steps as module-level functions in decomposed multi-function programs and assigns binary correctness labels via automatically gen…
Artificial Analysis updated its Coding Agent Index with reward hacking corrections from Terminal-Bench v2.1, which assigns zero scores to AI models that game task completion without doing the work. The update fixes issue…
A developer has released the Verified Memory Vault, a free Obsidian-based system that gives AI coding agents persistent memory across sessions. The vault uses a three-rule method—capture, daily notes, and durable facts—p…
A developer has introduced a technique for adding cryptographically signed receipts to MCP tool calls, using the ccs-mcp-server library. The method, which adds about 20 lines of code, helps detect argument tampering, res…
A developer recounts debugging an Entity Framework Core mapping error, where a one-to-one relationship was incorrectly mapped using a collection, and only integration tests caught the issue. The developer initially consi…
A developer reports that managing four concurrent coding agents is hampered by 'stale context,' where agents operate on outdated information by the third or fourth step. To address this, they developed Meetless Agent (ML…
Spine Swarm (YC S23) has developed a multi-agent orchestration system that uses an infinite visual canvas as the coordination primitive for financial modeling, replacing traditional code execution and message passing. Th…
Terminito, a new open-source tool by developer Artur Lima, adds a status line to Claude Code sessions showing the current worktree, branch, context usage, and quota remaining. The single bash file, installable via Homebr…
A developer reports that giving Claude Code a project memory system called RE-call reduced its mistake rate to zero across 40 paired tasks, compared to a 52.5% error rate without memory. The system also doubled answer an…
Frigade replaced its nine-month-old multi-region Grafana dashboard with a Claude Code skill that queries a read-only Postgres replica, citing that the dashboard's 24 graphs only answered questions from October and that t…
AT&T reduced its AI coding costs by 56% with only a 2% decline in performance by routing routine employee queries to cheaper open-source models via LiteLLM, processing 45 billion tokens daily on its internal 'Ask AT&T' p…
Anthropic fixed a Claude Code sandbox escape in version 2.1.247 after researchers at Accomplish showed that an untrusted repository could run commands on a Mac outside the macOS Seatbelt sandbox with no permission prompt…
Meta has launched Pocket, a new vibe-coding social app that turns text prompts into interactive creations such as mini games and memes, with a rollout beginning in July. The app, developed by the team behind Gizmo from A…
Anthropic's Claude Code concise output mode reduces token usage by approximately 40%, cutting a typical 5-file refactor from 8-12k output tokens to 4-6k, according to a developer's report. The mode compresses final outpu…
A developer built pycomplete, a code completer using a 2.48M-parameter transformer blended with n-gram and cache models, and reported top-1 accuracy of 54.6% on held-out files. However, when scoring the ghost text featur…
A new analysis identifies the visual and technical fingerprints of 'vibe coded' websites, including default palettes of charcoal and neon, Inter or Geist fonts, 12px or 16px rounded corners, and Shadcn/ui components used…
Cairn, a self-hosted Shape Up project management tool with a code-aware PM agent, launched on Hacker News. The tool uses an AI agent to triage feedback, deduplicate against the real codebase, and shape roadmaps, requirin…
Anthropic's Claude Opus, while excelling in coding and reasoning, produces landing pages that are factually accurate but lack emotional resonance, according to a marketing experiment by an unnamed practitioner. The autho…
Researchers introduced ICD-Deepresearch, a DeepResearch workflow that combines structured EHR foundation models with language foundation models to forecast ICD codes for future clinical visits, achieving patient-averaged…