Denoise · Twitter

The discourse has matured from theoretical jailbreaks to production security, with @AnthropicAI and @GoogleDeepMind establishing formal practices for agent safety.

Major AI labs are publishing formal frameworks and responsible disclosures related to red-teaming and securing autonomous agents.

• 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 primary interface for coding assistants is consolidating in the terminal, a shift signaled by @AnthropicAI's release and validated by commentary from @karpathy.

Anthropic's launch of Claude Code 1.5, a terminal-native coding agent, dominates the conversation, sparking discussion on developer experience shifts.

• 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 clear stack is forming: @OpenAI and @AnthropicAI define protocols, with infra providers like @vercel and @replit building the corresponding runtimes.

A wave of new tools for agent orchestration, deployment, and runtime hosting has been released by major players.

• 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 others focus on the agent stack, @MistralAI continues its strategy of releasing foundational open data assets to empower the broader research community.

MistralAI released a large-scale, open-source dataset for web-based optical character recognition (OCR).

• Mistral AI@MistralAIrising

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

A fault line is emerging between basic RAG and advanced context management, with voices like @GregKamradt questioning the former and tools like @mem0ai building the latter.

The discussion is shifting from simple RAG to more sophisticated 'context engineering' frameworks for managing memory in large-context models.

• 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.

A convergence pattern is visible: @NotionHQ and @linear add agent-like automation, while orchestration engine @temporalio is explicitly targeting agent workflows.

Workspace productivity tools are rolling out increasingly sophisticated automation features, blurring the line with agentic 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 industrializing, with platforms like @weights_biases providing quantitative analysis while practitioners like @dotey offer qualitative rules of thumb.

The focus is on scaling and standardizing prompt optimization, through large-scale benchmarking and sharing practical, battle-tested heuristics.

• 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 development matures, the focus shifts to subtle data quality issues, with @jerryjliu0 highlighting the specific problem of synthetic data poisoning.

A key challenge in training agents is filtering synthetic datasets to avoid data that harms model generalization.

• Jerry Liu@jerryjliu0repeated

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