Denoise · Twitter

The security conversation is maturing, with actors like @GoogleDeepMind and @MalwareTechBlog focusing on agent orchestration vulnerabilities, not just prompt injection.

Red-teaming frameworks and responsible disclosures for AI agents are being published by major labs like Anthropic and GoogleDeepMind.

• 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 tool space seems to be consolidating around terminal-native agents from players like Anthropic, signaling a potential shift away from IDE plugins.

Anthropic's release of Claude Code 1.5, a terminal-native agent, is the main event, with immediate benchmarks and adoption stories.

• 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 race is on to define the agent deployment standard, with OpenAI's SDK competing with more open approaches from LangChain and Temporal.

OpenAI, Vercel, and Replit released new infrastructure for deploying and orchestrating agents, while LangChain adapts to new protocols.

• 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 conversation, MistralAI continues to strengthen its position in open-source AI by providing foundational multimodal datasets.

MistralAI released a large-scale, open OCR dataset for training multimodal models.

• Mistral AI@MistralAIrising

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

Practitioners like @GregKamradt are arguing for a move beyond vector retrieval, suggesting a convergence towards complex memory architectures from @mem0ai and @llamaindex.

The conversation is shifting from simple RAG to more complex "context engineering" and layered memory systems for agents.

• 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 patterns of agentic orchestration seen in dev tools are now appearing in general productivity software from @NotionHQ and @linear.

Workspace automation tools like Notion and Linear are adding AI-driven features for auto-filling and triaging tasks.

• 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.

Prompt engineering is becoming more scientific, with firms like @weights_biases advocating for large-scale, data-driven optimization over ad-hoc tuning.

Discussion centers on systematic benchmarking of system prompts and practical tricks learned from production usage.

• 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 articulated by @jerryjliu0, the focus in agent training is shifting from raw data quantity to the quality of synthetic data curation.

A discussion on the importance of curating high-quality synthetic data for training agents to avoid poisoning generalization.

• Jerry Liu@jerryjliu0repeated

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