Denoise · Twitter

The threat model is shifting from model-level prompt injection (Anthropic) to vulnerabilities in the agent orchestration layer (Google DeepMind).

Attention is centered on red-teaming autonomous agents, with major labs releasing frameworks and researchers demonstrating real-world pentesting.

• 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 battle for developer mindshare is moving from IDE plugins (Copilot) to standalone terminal agents (Claude Code), with early benchmarks from @swyx beginning to differentiate them.

Major releases from Anthropic and DSPy, amplified by voices like Karpathy, confirm a structural shift towards terminal-native coding agents.

• 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 de-facto agent stack is converging around OpenAI's SDK and Vercel/Replit's serverless worker model, with LangChain providing the interoperability layer.

Infrastructure providers are racing to support the new agent paradigm with SDKs, deployment harnesses, and dedicated runtimes.

• 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, foundation model players like MistralAI continue to build core data moats for vision and document understanding.

MistralAI contributes a massive, cleaned web OCR dataset to the community 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 replaced by more specific concepts from @GregKamradt and @mem0ai, signaling a maturation of the field beyond basic vector retrieval.

The discourse is evolving past simple RAG to 'context engineering,' focusing on complex memory architectures and large-context failure modes.

• 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 agent paradigm is being integrated into mainstream SaaS products like Notion and Linear as a core automation and productivity feature.

Workspace automation tools like Notion and Linear are shipping agent-like features, while Temporal offers durable execution primitives for complex workflows.

• 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 a data-driven discipline, with platforms like Weights & Biases enabling industrial-scale testing and frameworks like DSPy automating optimization.

The focus is shifting from anecdotal tricks to systematic, large-scale benchmarking and automated optimization 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 points to a key bottleneck for agent improvement: filtering synthetic training data to avoid poisoning generalization without clear metrics.

A single signal highlights the critical but overlooked challenge of curating high-quality synthetic data for training effective agents.

• Jerry Liu@jerryjliu0repeated

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