Denoise · Twitter

The critical attack surface is shifting from the LLM itself to the agent's orchestration layer, a pattern identified by Anthropic, DeepMind, and @AlexAlbert__.

Major labs are publicly dissecting agent security, focusing on red-teaming complex jailbreaks and novel attack surfaces 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 developer tool landscape is seeing a split between IDE-integrated assistants like Copilot and standalone terminal agents like Claude Code, with early adopters like @levelsio reporting productivity gains.

Anthropic's Claude Code 1.5 launch signals a direct move into terminal-native agents, prompting immediate benchmarks and developer workflow changes.

• 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 convergence is happening around a standardized stack for agent deployment. Offerings from OpenAI, Vercel, and Replit are creating composable primitives for building and hosting durable agents.

OpenAI, Vercel, and Replit released new infrastructure for deploying and managing autonomous agents, focusing on orchestration, tool-calling, and background workers.

• 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.

Despite the focus on agents, the release of high-quality, open datasets by major players like Mistral AI remains a critical driver for progress in the open-source modeling community.

Mistral AI released a large-scale, cleaned web OCR dataset, providing a foundational resource for training multimodal models.

• Mistral AI@MistralAIrising

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

A consensus is forming that vector search alone is insufficient for agent memory. New approaches from @mem0ai and @GregKamradt distinguish between working memory and long-term stores.

The discourse is moving beyond simple RAG, with developers exploring complex memory architectures and 'context engineering' to handle long-running agent tasks.

• 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.

Traditional SaaS platforms are building deterministic, workflow-based automation that competes directly with the value proposition of less reliable LLM-based agents for business process tasks.

Workspace tools like Notion and Linear are shipping significant non-AI automation features, improving productivity through chained database updates and auto-triaging.

• 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 optimization is industrializing, moving from individual 'tricks' (@dotey) to statistically-driven methods for finding the most effective system prompts at scale (@weights_biases).

Prompt engineering is becoming more systematic, with large-scale benchmark reports from firms like Weights & Biases analyzing thousands of variations.

• 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 complexity grows, the challenge circles back to a classic ML problem: curating high-quality training data. @jerryjliu0 highlights the difficulty of identifying synthetic data that poisons generalization.

The focus in agent training data is on sophisticated filtering of synthetic data to avoid performance degradation and generalization failures.

• Jerry Liu@jerryjliu0repeated

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