Denoise · Twitter

The security conversation has shifted from chatbot vulnerabilities to systemic risks in agent orchestration, with both Anthropic and GoogleDeepMind focusing on tool interaction and multi-step execution.

Major AI labs are now publicly disclosing red team frameworks and patched jailbreaks for autonomous agents, moving beyond simple prompt injection.

• 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.

The primary competition in AI coding is moving from IDE plugins (Copilot) to standalone terminal agents, with Claude Code and Codex now being benchmarked on this new interaction model.

Anthropic's release of Claude Code 1.5 as a terminal-native agent dominates the conversation, framed by figures like @karpathy as a paradigm shift away from IDEs.

• 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.

A clear convergence is forming around a standardized agent stack: OpenAI is providing the core protocol, while platforms like Vercel and Replit are racing to provide the best hosting and orchestration layer.

OpenAI, Vercel, and Replit all released new infrastructure for agent deployment, including SDKs, edge runtimes, and one-command deployment harnesses.

• 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.

While agents dominate the discourse, foundational data releases like Mistral's are a critical, less visible layer enabling the next generation of models that will power these agents.

Mistral AI released a 100M-row web OCR dataset, providing a foundational asset for training future multimodal models.

• Mistral AI@MistralAIrising

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

A consensus is emerging that simple vector retrieval is insufficient for stateful agents. @GregKamradt, @mem0ai, and @llamaindex are all pointing towards more structured, multi-layered memory systems.

The conversation is evolving from RAG to 'context engineering,' focusing on complex memory architectures and failure modes in very large context windows.

• 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.

The agentic pattern is being productized for non-technical users. While dev tools focus on terminal agents, SaaS companies like Notion and Linear are embedding similar autonomous logic directly into workflows.

Workspace tools like Notion and Linear are shipping autonomous features, such as auto-populating tables and triaging issues, reflecting the broader agent trend.

• 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 field is bifurcating between artisanal prompt crafting (@dotey) and industrial-scale, automated optimization (@weights_biases, @dspy_ai), suggesting a specialization of skills is emerging.

Prompt engineering is maturing into a systematic discipline, with large-scale benchmarks from @weights_biases contrasting with tactical advice from practitioners like @dotey.

• 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.

As agent training becomes more accessible, the bottleneck is shifting from raw compute to sophisticated data pipelines, a problem highlighted by practitioners like @jerryjliu0.

The key infrastructure challenge for training agents is now seen as data curation, specifically filtering synthetic data to avoid generalization poisoning.

• Jerry Liu@jerryjliu0repeated

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