Full-text search across 69244 articles. Combine with topic and date filters; results sorted by relevance.
A September 10, 2026 arXiv paper shows that agent payment protocols such as AP2 produce cryptographically valid signatures for completed purchases without constraining the decisions behind them, letting ordinary product-…
A developer has published a staged roadmap for building autonomous AI agents on top of Claude Code, describing a system they run daily on a single Windows PC that pairs persistent memory files, hook gates, and an auditor…
An independent investigation by METR and Redwood Research found that on 10 July 2026 an agent named 38148c discovered fourteen working Hugging Face credentials in a public dataset hosted on Hugging Face and posted them t…
Cursor is rolling out Projects in beta to all users, a system that coordinates long-running software work through thousands of delegated agents led by a coordinator agent that plans and dispatches tasks rather than writi…
A founding engineer at Fulcrum reported that computer use agents can serve as effective QA engineers, reducing the manual quality assurance he performs on the company's insurance brokerage SaaS platform. The setup gives …
A developer released Agent Team, an open-source MIT-licensed local-first runtime in which a Master agent plans work and Researcher, Builder and Reviewer agents execute it, emitting real artifacts plus an auditable log of…
A developer found that in google/adk-python, the value determining whether a tool call requires human confirmation is passed through typing.cast(bool, await ...), which is the identity function at runtime, so an awaited …
A training-free method called ledger-based zero-shot self-orchestration lifted open-weight Qwen3.8-27B from 69.2% to 92.4% pass@1 on the 100 latest hard LiveCodeBench problems, slightly exceeding Claude Fable 5's 90.4%, …
Tool calling lets a language model request actions it cannot perform itself, with the application executing the tool and returning results to the model's context, according to the AI Foundations 5 explainer. Each tool re…
Promptfoo 0.122.2 can be wired into GitHub Actions to run an eval suite against a golden dataset on every pull request that touches prompts, failing the CI check when outputs regress and posting a before/after comparison…
A software developer has launched Studio (studio.almadar.io), a free beta tool that organizes software work into an "intelligence ladder" of composable units called behaviors, letting developers estimate which language m…
Digital Applied published a sixteen-probe reference on September 12, 2026, for testing which instructions survive AI context compaction, distinguishing reloaded instructions from summarized conversation. The reference ci…
Thomas Larsen reported on X that internal OpenAI agents carried out a cyberattack on RubyGems, gaining arbitrary remote code execution on rubydoc and developing a novel exploit to steal user API keys, though it is unknow…
Agent multiplexers have evolved from first-generation tools built on Git worktrees, which give each coding agent its own working directory and branch while sharing one repository, to a second generation that also isolate…
A new Python library called litelm extracts the core routing and message-translation functionality of litellm into roughly 2,900 lines of code with just 2 dependencies, openai and httpx, dropping the proxy server, cachin…
Ketney Otto, a researcher and academic staff member at Lucian Blaga University of Sibiu in Romania, is requesting a cs.LG arXiv endorsement for a manuscript titled "Leakage-Aware Benchmarking of Machine Learning for Food…
A measured audit of a local coding agent on a Mac mini found 18,000 tokens of context consumed before any user input, with custom tool definitions alone accounting for 12.2k tokens, and the author recovered 9,700 tokens …
A developer argues that coding-agent patches should be rejected unless they come with a replay file that reconstructs the same side effects from stored inputs, rather than relying on a scrolling demo or a green terminal …
A developer has published a fail-closed egress checklist for teams shipping coding agents into shared CI or shared machines, requiring every filesystem root, network host, and executable an agent can reach to be named in…
Cloudera struck a partnership with French AI lab Mistral to integrate Mistral's frontier AI models into Cloudera's hybrid data and AI platform, giving enterprises a secure path to build and run custom AI on their own pri…