Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
The proxy delivers simultaneous token cost reduction and accuracy improvement over plain JSON — without requiring any changes to existing MCP servers — by replacing a format that causes LLM comprehension failures at scale with one that scores 90.7% vs. JSON's 53.6% on the same data.
The rebuilt scoring model replaces a system that compressed 85.7% of tools into a single grade, giving the ecosystem its first meaningful quality differentiation signal for identifying which MCP servers are actually discoverable by AI agents.
MSA demonstrates that a 109B-parameter model can process 1M-token contexts with 28.4x less attention compute and 14.2x faster prefill, making million-token agentic and code-reasoning workloads substantially more feasible at deployment scale.
The tool directly addresses a concrete bottleneck in agentic coding loops — context budgets consumed by redundant file re-reads — by fitting entire repositories into context that previously only held a fraction of the codebase.
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.
`brooks-lint` directly addresses a gap where AI-generated code passes functional tests but violates established architectural principles — by encoding those principles from classic texts into a reusable review skill, it applies structured software-engineering judgment to AI-written codebases.
The conference program shows that the AI coding stack debate has shifted from "should we do context engineering" to harder second-order problems — skill sprawl, supply chain security, and harness design — marking a concrete maturation in how the industry frames agentic development.
TrajGenAgent demonstrates that a fine-tuning-free, hierarchical agent design can match or exceed the trajectory realism of computationally expensive fine-tuned models, lowering the barrier to generating privacy-safe synthetic mobility data for transportation, urban planning, and epidemic control applications.
The hook replaces a probabilistic `CLAUDE.md` suggestion — which the model could rationalize past — with a hard, pre-execution wall that reduces `--no-verify` bypasses from one-in-five to zero, demonstrating how `PreToolUse` hooks can enforce truly non-negotiable constraints on agentic behavior.
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.