Denoise · Twitter

The focus on agent security from @AnthropicAI and @GoogleDeepMind signals a maturity shift, treating agent vulnerabilities as a formal discipline, not an afterthought.

Major labs are publishing formal red-teaming frameworks and responsible disclosures for agent-based 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.

The narrative, reinforced by @karpathy and @swyx, consolidates around the agent-in-terminal as a new developer UX paradigm challenging IDE-integrated tools.

Anthropic's launch of a terminal-native coding agent, Claude Code 1.5, dominates the conversation, with benchmarks and developer adoption testimonials.

• 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 split in strategy is visible: @OpenAI offers low-level SDK primitives for an ecosystem, while @vercel and @replit provide higher-level, integrated deployment solutions.

Foundational infrastructure for agents is being released, including SDKs for orchestration from OpenAI and specialized deployment runtimes from Vercel and Replit.

• 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.

Quiet day. The release from @MistralAI is a standard contribution, but lacks a connection to the dominant agent narrative of the day.

A single large-scale 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 appears between the theoretical promise of massive context windows (@reach_vb) and the practical complexity of using them effectively (@GregKamradt).

Practitioners are debating the limits of large context windows and traditional RAG, proposing new 'context engineering' frameworks.

• 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' of SaaS is happening quietly, with @NotionHQ and @linear shipping agent-like automation features without using explicit AI branding.

General workspace automation features are being released by Notion and Linear, alongside discussion of durable execution for agents from Temporal.

• 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 discussion moves from anecdotal prompt 'tricks' (@dotey) to systematic, data-driven optimization (@weights_biases), reflecting a maturing engineering discipline.

Focus is on empirical, large-scale benchmarking of system prompts to find efficient frontiers.

• 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.

This lone signal from @jerryjliu0 points to a critical but less-discussed bottleneck in the agent development cycle: data quality over data quantity.

A single tweet discusses the challenges of curating synthetic data for training agents.

• Jerry Liu@jerryjliu0repeated

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