Denoise · Twitter

Both Anthropic and Google DeepMind are publicly releasing frameworks and disclosures, indicating a push to standardize agent security practices before capabilities escalate further.

Major AI labs and independent researchers are converging on methodologies for red-teaming autonomous agents, moving 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 @AnthropicAI and amplified by @karpathy and @swyx, reframes the developer tool competition away from IDE plugins towards standalone, conversational agents.

The launch of Anthropic's Claude Code 1.5 marks a significant push toward terminal-native agents as the primary developer interface.

• 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, Replit, and LangChain are all converging on providing developers with primitives for multi-worker, durable agent deployments, creating a new infrastructure layer.

The ecosystem is rapidly building standardized infrastructure for deploying, orchestrating, and hosting autonomous agents.

• 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.

MistralAI continues its pattern of contributing foundational artifacts (datasets, open models) to the community, even as others focus on agent applications.

A quiet day, with the only notable signal being a large-scale open dataset release for web OCR from MistralAI.

• Mistral AI@MistralAIrising

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

A fault line is emerging between scaling context windows (@reach_vb) and developing sophisticated 'context engineering' frameworks (@GregKamradt, @mem0ai), questioning if one can replace the other.

Practitioners are exploring the failure modes of massive context windows and moving beyond simple RAG to more 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 trend mirrors the agentification seen in development tools, suggesting a broader convergence on using autonomous systems to manage structured business workflows.

Workspace automation is a recurring theme, with tools like Notion and Linear adding agent-like capabilities to manage information and triage 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.

Efforts by @weights_biases exemplify a move towards treating prompt optimization as a rigorous engineering discipline, rather than an intuitive art.

The focus in prompt engineering is shifting from anecdotal tricks to scalable, systematic benchmarking of system prompts against multiple models.

• 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 training becomes more common, @jerryjliu0's focus on filtering out data that 'poisons generalization' points to a new set of data quality challenges beyond standard supervised learning.

A niche but critical signal highlights the importance 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.