Denoise · Twitter

Anthropic and Google DeepMind are standardizing how to audit autonomous agents, focusing on complex exploits beyond simple prompt injection.

Major labs are publicly disclosing agent vulnerabilities and releasing formal red-teaming frameworks, shifting security from theory to practice.

• 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 conversation, led by figures like karpathy and swyx, is coalescing around the terminal agent as a fundamental shift in developer workflow, moving past IDE plugins.

The release of Anthropic's Claude Code 1.5 has triggered immediate adoption and benchmarking against established tools like Copilot.

• 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.

Multiple players are racing to build the agent stack: OpenAI is providing the SDK, LangChain is standardizing protocols, and Vercel/Replit are offering managed runtimes.

A wave of new releases from OpenAI, Vercel, and Replit provides the primitives for orchestrating and deploying autonomous agents at scale.

• 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 a significant data contribution, this release is a foundational infrastructure move, with most of the day's developer activity focused on agentic workflows.

MistralAI released a large-scale OCR dataset, providing foundational material for training new multimodal models.

• Mistral AI@MistralAIrising

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

A consensus is emerging, articulated by GregKamradt and mem0ai, that vector search alone is an insufficient memory model for sophisticated agents.

The discussion around providing context to models is evolving from simple RAG to more complex 'context engineering' and structured memory systems.

• 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 agentification trend is not limited to developer tools; business SaaS platforms like Notion and Linear are building their own versions of autonomous task execution.

Productivity tools like Notion and Linear are shipping workspace automation features that function as specialized, embedded agents.

• 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.

Weights & Biases and DSPy are pushing to transform prompt creation from a manual art into a data-driven optimization problem.

Prompt engineering is becoming more systematic, with large-scale benchmarking efforts and automated optimization techniques gaining traction.

• 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 architectures mature, Jerry Liu points out that the bottleneck is returning to data quality, specifically filtering out synthetic data that harms generalization.

The focus in agent training is shifting towards sophisticated data curation to avoid performance degradation from low-quality synthetic data.

• Jerry Liu@jerryjliu0repeated

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