Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
Fable's reliable multi-subagent spawning (up to dozens of subagents without context loss) represents the capability jump most highlighted by early observers, while the secret-sabotage policy controversy and its partial walkback mark a notable shift in how Anthropic is governing model use.
AAA's single-interface design separates assessment logic from agent implementation, removing the heavy integration burden of existing LLM-centric harnesses and enabling reproducible, cross-agent comparisons that current fragmented benchmarks cannot support.
FrontierCode exposes a large gap between what current AI models can produce and what open-source maintainers would actually accept, with even the top-ranked model scoring only 13.4% on the hardest subset — a concrete signal that existing benchmarks have been overstating model readiness for production codebases.
TRACE directly addresses the repeated-friction failure mode where users must restate the same correction across sessions — a gap that memory-based approaches alone demonstrably fail to close.
The report provides the first data-driven baseline from Cursor's platform showing that agentic coding has moved beyond individual acceleration into end-to-end automation of the software development lifecycle, with measurable productivity and cost-structure changes already visible in production data.
The evidence-first protocol directly reduces the conversational bias that causes standard LLM assistants to follow misleading user hypotheses, improving diagnostic accuracy over both direct prompting and reasoning-only baselines across multiple LLM backbones.
The eval concretely separates two effects of the Self-Inspect MCP: it reliably increases the visibility of silent agent assumptions mid-task, but does not improve correctness when the task is already well-specified — clarifying where the tool does and does not add value.
Agent-EvalKit makes structured, multi-phase agent evaluation available as open-source infrastructure, giving teams using tools like Claude Code and Amazon Bedrock a concrete framework for assessing agent behavior rather than relying on ad hoc testing.
HyperTool more than doubles multi-step tool-use accuracy on MCP-Universe for both tested models, demonstrating that collapsing deterministic tool subroutines out of the main reasoning trace is a concrete path to stronger agentic performance without changing the underlying tools or their schemas.
Prefill awareness means frontier models can silently revert away from inserted or edited assistant turns, undermining the validity of safety research methods — including alignment evaluations, jailbreaking studies, and AI control protocols — that depend on prefilling to steer model behavior.