Denoise · Twitter

The focus is shifting from prompt-level exploits to orchestration-level vulnerabilities, with @GoogleDeepMind providing frameworks and @AnthropicAI disclosing real-world patches.

The security conversation is now centered on proactive red teaming and vulnerability disclosure for autonomous agents.

• Anthropic@AnthropicAIrising

  Responsible disclosure on a Claude jailbreak chain we patched last week. Full write-up including our red team timeline.

• Google DeepMind@GoogleDeepMindrising

  New red team framework for prompt injection in autonomous agents. Covers cross-tool leakage, scanner evasion, and sandbox escape patterns.

• MalwareTech@MalwareTechBlogrepeated

  Autonomous agent running pentest flows against a real SaaS. First real-world run: fewer false positives than I expected on the vulnerability surface.

A convergence on the terminal as the primary UI for coding agents is evident, with @AnthropicAI's Claude Code release and @karpathy's analysis leading the narrative.

Major players are shipping terminal-native coding agents, signaling a potential shift away from IDE-centric development workflows.

• Anthropic@AnthropicAIrising

  Claude Code 1.5 is live. Terminal-native coding agent with full Claude Opus reasoning, file-ops sandbox, and session replay.

• Andrej Karpathy@karpathyrising

  The developer-experience shift from IDE to terminal agent is underrated. Coding workflows are about to look nothing like 2024.

• swyx@swyxrising

  Codex vs Claude Code terminal agent benchmarks. Pass@1 diverges more than I expected on the long-context editor tasks.

• DSPy@dspy_airising

  DSPy 3.0: prompt optimization via compile-time search over system prompt variations. Benchmarks inside.

• @levelsio@levelsiorising

  Switched my whole editor setup to Claude Code this week. Shipping faster than when I used Cursor + Copilot.

The entire stack is converging, from @OpenAI's protocol-level SDKs to @vercel and @replit's dedicated agent runtimes, with @LangChainAI providing integration.

A new infrastructure stack is rapidly forming to support the deployment and orchestration of autonomous agents.

• OpenAI@OpenAIrising

  New agent SDK: protocol-level tool calling, deployment harness, and multi-worker orchestration primitives. Docs live.

• LangChain@LangChainAIrising

  MCP protocol integration thread. How to wire existing LangGraph agents into the Anthropic Model Context Protocol server spec.

• Vercel@vercelrising

  Edge runtime for agent workers is live. Spawn durable background agents from any serverless deployment.

• Alex Albert@AlexAlbert__rising

  When your security scanner finds nothing scary on an agent deploy, check the orchestration layer again. That's usually where the jailbreak sneaks through.

• Replit@replitrising

  New agent deployment harness. One command to go from local orchestration to hosted agent worker.

@MistralAI's dataset release indicates continued investment in foundational multimodal capabilities, even as the agent narrative dominates the discourse.

A large, clean dataset for web OCR has been open-sourced, providing a key resource for training multimodal models.

• Mistral AI@MistralAIrising

  Open dataset release: 100M-row web OCR dataset. Cleaned, licensed, ready to train.

The term RAG is being fragmented into more specialized concepts, with @GregKamradt advocating for 'context engineering' and @mem0ai proposing tiered memory architectures.

The conversation around providing models with external knowledge is evolving from simple RAG to more complex 'context engineering' frameworks.

• Vaibhav Srivastav@reach_vbrising

  Tested the new 10M context memory window end to end. Surprising failure modes around rag retrieval cache invalidation, thread below.

• Greg Kamradt@GregKamradtrising

  RAG is dead, long live context engineering. My framework for when to cache, when to retrieve, and when to just dump memory into the prompt.

• mem0@mem0airising

  Memory layer for agents: differentiating working memory from the subconscious store. Vector index isn't enough anymore.

• LlamaIndex@llamaindexrepeated

  Knowledge graph retrieval walkthrough: when semantic vector search misses, graph hop beats it every time.

While the AI frontier pushes agents, established SaaS tools like @NotionHQ and @linear are finding value in more predictable, chained automation for core workflows.

Major workspace platforms are rolling out non-agentic, rule-based automation features to improve productivity.

• Notion@NotionHQrising

  Notion workspace automation is out of beta. Auto-fill tables, chained updates across databases, and a new audit log surface.

• Linear@linearrising

  Linear now auto-triages incoming issues. Quiet launch, but already our favorite workspace feature of the year.

• Temporal@temporaliorepeated

  Orchestrating agents with durable workflows: replayable, resumable, and multi-worker by default. Walkthrough from our infra team.

• James Clear@jamesclearrepeated

  The best habit tracker is the one you actually open. Three open-source alternatives worth trying.

The practice is maturing from anecdotal tricks (@dotey) to large-scale empirical analysis, exemplified by @weights_biases' 40k variant benchmark study.

The focus in prompt engineering is shifting towards scalable, data-driven benchmarking of system prompts.

• dotey@doteyrising

  Five prompt tricks learned this week from reviewing 200 production prompts. Short thread.

• Weights & Biases@weights_biasesrising

  System prompt benchmarking at scale: we ran 40k variants across 6 frontier models. The efficient frontier is not where you think.

@jerryjliu0's analysis points to data poisoning as a key risk, suggesting that data filtering is becoming as important as model architecture for agent development.

The quality of synthetic data curation is highlighted as a critical factor for training capable and generalizable agents.

• Jerry Liu@jerryjliu0repeated

  Dataset curation for agent training: how we filter synthetic data that looks good but poisons generalization.