Denoise · Twitter

The security conversation is rapidly maturing from prompt injection to orchestration-layer vulnerabilities, with both @AnthropicAI and @GoogleDeepMind focusing on how agents misuse tools.

Frontier labs are disclosing agent jailbreaks and publishing formal red-teaming frameworks for agentic systems.

• 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.

There is a clear convergence on the terminal as the new IDE, a trend championed by @AnthropicAI's product and articulated as a major UX shift by @karpathy.

Anthropic's release of Claude Code 1.5, a terminal-native agent, has triggered widespread discussion about a fundamental shift in developer workflows.

• 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 standards race for agent orchestration is beginning, with @OpenAI and @replit building specific deployment tooling that competes for developer adoption against more general platforms.

Major players shipped agent-native infrastructure, including SDKs for orchestration, deployment harnesses, and dedicated edge 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.

This category is quiet, with @MistralAI's dataset release representing an input-side contribution rather than a new on-device technique or multimodal architecture.

Mistral AI released a large, open-source web OCR dataset for training foundational models.

• Mistral AI@MistralAIrising

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

The simple vector database approach to RAG is being challenged, with @GregKamradt, @mem0ai, and LlamaIndex all proposing more structured, multi-layered memory solutions for agents.

The conversation is evolving from simple RAG to 'context engineering,' focusing on complex memory architectures and caching strategies.

• 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 pattern of 'workspace-as-agent' emerges, where tools like @NotionHQ and @linear are automating user workflows, reflecting a broader trend towards autonomous systems.

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

• 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 craft of prompt engineering is professionalizing, with platforms like @weights_biases moving the practice from anecdotal advice to data-driven optimization.

Attention is shifting from individual prompt tricks to systematic, large-scale benchmarking of system prompts to find optimal configurations.

• 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.

A quiet day for this category, with @jerryjliu0's post highlighting the subtle but critical challenge of data quality in the agent training pipeline.

A single signal focused on the nuances of dataset curation for agent training, specifically filtering harmful synthetic data.

• Jerry Liu@jerryjliu0repeated

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