Denoise · Twitter

The agent security conversation is rapidly maturing from theoretical to applied, with @AnthropicAI and @GoogleDeepMind converging on transparent, framework-driven security practices.

Major labs are publicly disclosing agent jailbreaks and releasing formal red-teaming frameworks.

• 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 developer workflow is consolidating around the terminal, with @AnthropicAI's product launch and @karpathy's commentary signaling a potential paradigm shift away from traditional IDEs.

Anthropic's release of a terminal-native coding agent, Claude Code 1.5, dominates the discussion, with benchmarks and developer testimonials.

• 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.

OpenAI, Vercel, and Replit are converging on providing the foundational PaaS layer for agents, abstracting away orchestration and worker management.

A wave of new infrastructure for agents has been released, including an SDK from OpenAI and deployment harnesses from Vercel and Replit.

• 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.

It's a quiet day focused on data primitives, with @MistralAI providing a foundational asset for training rather than a new model or technique.

The only signal today is a large-scale web OCR dataset released by MistralAI.

• Mistral AI@MistralAIrising

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

Practitioners like @GregKamradt and tools like @mem0ai and @llamaindex are converging on the idea that simple vector search is insufficient for advanced agents.

The conversation moves past simple RAG towards 'context engineering,' sophisticated memory architectures, and knowledge graph retrieval.

• 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.

@NotionHQ and @linear are converging on a pattern of using AI for structured task automation, a parallel trend to the open-ended agent development in other categories.

Workspace automation tools like Notion and Linear are launching AI-powered features for auto-filling and triaging structured data.

• 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 of prompt engineering is professionalizing, with firms like @weights_biases demonstrating a data-driven approach that refutes reliance on simple heuristics.

The focus is on moving from anecdotal prompt tricks to systematic, large-scale benchmarking to find optimal 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 points to a critical bottleneck in agent development: the quality and filtering of synthetic training data, not just its quantity.

A discussion highlights the importance of curating high-quality synthetic data for training agents to avoid performance degradation.

• Jerry Liu@jerryjliu0repeated

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