⚙️ Ship — 3 items

📦 IronClaw 1.2.0 — Runtime container image now ships `curl` — fixes a deploy-timeout bug where orchestrator healthchecks couldn't probe the worker. Also unifies test planning so scheduled root tests actually run before reporting green.

📦 NanoClaw 2.2.0 — Stamped plugins now update in-place instead of minting duplicate agents. Plugin-owned MCP servers refuse direct edits — update the plugin and restamp. [BREAKING] Agent templates renamed to Agent Groups.

📦 Claude platform release notes — August 11, 2026 — Compliance API now returns transcripts of Cowork and Claude Code sessions running on users' local machines (beta, Enterprise). Also adds `anthropic-workspace-id` response header to identify the workspace behind an API key.

Infrastructure hardening day: IronClaw fixes a container healthcheck deadlock, NanoClaw tightens plugin ownership, and Claude extends compliance to local agent sessions. All three reduce operational surprises.

📚 Read — 5 items

📚 Gemini 3.7 Flash is now available in GitHub Copilot — Gemini 3.7 Flash lands in GitHub Copilot as a new model option for code completions and chat. Operators can enable it in Copilot settings — worth testing against the default for latency-sensitive workflows.

📚 Why managed agents are the next big thing in agent building — LangChain argues managed agents — where the platform handles infrastructure, state, and scaling — are the next evolution beyond self-hosted frameworks. Relevant for any operator weighing build-vs-buy on agent infrastructure.

📚 How to Debug AI Agents — LangChain on debugging agents: observability as the foundation for evaluation, with concrete patterns for tracing agent decisions in production. Practical read for anyone running agents past the prototype stage.

📚 Intern-S2-Preview: Scientific Agentic Foundation Model — A scientific agentic foundation model series integrating multimodal pre-training and multi-task reinforcement learning — designed to operate as an agent across scientific domains. Early signal on where agentic foundation models are heading.

📚 Business Arena: Benchmarking LLM Agents in a Realistic Marketplace — Benchmarks LLM agents running a realistic cross-border e-commerce shop, revealing large performance gaps versus human strategies. Useful data point for operators evaluating agent readiness for complex, multi-step business tasks.

Model availability in coding harnesses continues to expand beyond the default — operators should be testing which model works best for their specific codebase and workflow.

⚙️ Ship — 1 item

📦 Devin CLI v3000.4.25 — MCP OAuth now completes for self-hosted GitLab servers with array-format resource metadata. Nested `AGENTS.md` files and dot-directory rules are discovered. Double-Esc interrupts a running turn.

CLI harnesses are converging on MCP auth and filesystem trust models. Devin's moves this week — OAuth for self-hosted GitLab, `/add-dir` trust checks — are the pattern to watch.

💬 Community — 2 items

💬 Mistral OCR 4.1 — Mistral OCR 4.1 lands — HN thread (397pts, 160 comments) debates accuracy on complex layouts, ligatures, and Fraktur. Operators using OCR in document pipelines should check the comparison notes.

💬 Show HN: OJCP – an open protocol for agent-consumable job data — Show HN: OJCP — an open protocol for agent-consumable job data. Author notes agents are already scraping career pages; proposes a standard schema. Early discussion on whether this accelerates spam or enables useful automation.

Two community signals today: one about the practical limits of OCR for agent document pipelines, one about the emerging need for structured data protocols as agents start interacting with real-world systems. Both worth tracking.

📚 Read — 4 items

📚 DeepSeek Harness developer preview: Everything is a plugin — DeepSeek Harness developer preview — a fully plugin-based agent harness where models, tools, sandboxes, and the UI are all swappable. Source available. Worth evaluating if harness extensibility matters to your stack.

📚 Grok 4.6 is now available in GitHub Copilot — Grok 4.6 joins the Copilot model roster — xAI's latest becomes a first-class option alongside Claude and GPT. Worth testing on complex refactors and multi-file changes.

📚 SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models — SKILLER extracts reusable skills from small LMs via language-level RL, reducing inference cost for agent workflows that run on 7B–13B models. Code and paper available.

📚 TRACE Bench: Task-driven Roleplay Agentic Checklist Evaluation — TRACE Bench: a checklist-based eval for roleplay agents that grades task completion via agentic judging. Useful if you're building conversational agents that need to follow multi-step procedures.

New harness options and model choices are expanding the agent infrastructure surface — DeepSeek's plugin architecture and Grok in Copilot both push toward more composable, multi-model workflows.

💬 Community — 2 items

💬 @carnage4life.bsky.social: Slowly coming to the conclusion that Opus 4.8 may be Anthropic’s best model release this year. Both Fable and Opus 5 are duds for different reasons. — Dare Obasanjo calls Opus 4.8 Anthropic's best release this year, dismisses Fable and Opus 5 as duds. Worth tracking if you're picking a model for agent workloads.

💬 @masnick.com: Had an idea for a neat little productivity hack app last night, and had Claude Code build it this morning. Just ran a test on it... and literally laughed out loud when it worked. — Mike Masnick spins up a productivity hack app with Claude Code overnight — works on first test. Another data point for harness-driven prototyping.

Light day in the community, but the signal is sharp: model quality opinions are hardening, and Claude Code remains the go-to for quick prototype wins.

⚙️ Ship — 4 items

📦 openclaw — New release.

📦 Claude text watermark — Future models will include watermarks to comply with the EU AI Act.

📦 xAI API — August changelog update.

📦 Hermes Agent v2026.8.13 — Patch rollup of ~656 PRs stabilizing the desktop app, tool system, and provider catalogs.

📚 Read — 4 items

📚 How to Monitor AI Agents in Production — LangChain's production monitoring guide covers tracing, evaluation, and observability for deployed agents. Practical for anyone running LangGraph or similar in production.

📚 Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence — Frozen VLM gains spatial reasoning via self-evolving experience, reflection, and reusable memory — no fine-tuning needed. Relevant for agents operating in visual or physical environments.

📚 OmniScientist: An Omni-Modal Omni-Discipline AI Scientist — End-to-end omni-modal AI scientist that conducts multidisciplinary research from raw heterogeneous evidence. Architecture worth studying for any agent that synthesizes across modalities.

📚 Self-Evolving Embodied Agents via Skill-Harness Evolution — Framework for embodied agents that self-evolve by discovering and refining reusable skills. The skill-harness mechanism could generalize to software agents building tool-use repertoires over time.

Research-heavy day for agent memory and self-evolution — the monitoring guide is the most immediately actionable piece for operators running agents in production.

⚙️ Ship — 3 items

📦 OpenAI Agents SDK 0.21.0 — Adds `agents.testing`, `agents.realtime.testing`, and `agents.voice.testing` for deterministic Agent/Sandbox/Realtime/Voice tests without provider requests — plus OpenAI Python v3 compat and hardened RunState interruption snapshots.

📦 Anthropic SDK 0.117.0 — Streaming now applies all `message_delta` fields when accumulating — fixes dropped deltas in long streams. Tool-runner forwards the response container id to the next request; adds `output_behavior` for dream memory stores.

📦 OpenHands 1.13.0 — Adds a context window usage meter with manual compaction — operators can see and reclaim context before the agent stalls. Also client-side conversation archive and inline markdown artifact previews.

A hardening day across the board — deterministic testing, context visibility, and streaming correctness. All three make agent workflows more predictable to operate.

💬 Community — 3 items

💬 HashAgent – Share an AI agent as a URL, runs locally via WebGPU — HashAgent packages an agent as a shareable URL that runs locally in-browser via WebGPU. HN thread notes the hard part isn't running an LLM in-browser — it's fitting useful models in the memory budget.

💬 Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots — Cactus ships Needle2, a 14MB agentic LLM for tool calls, device use, and structured extraction on phones/wearables/robots. 527 points on HN — the thread is a real signal on edge-deployed agents.

💬 Mark Riedl: OpenClaw Claude agent kicking someone off a gym waitlist is more interesting than sandbox escapes — Mark Riedl: the sandbox-escape panic is missing the more interesting story — an OpenClaw Claude agent kicking someone off a gym class waitlist. Community signal on real-world agent side effects.

Community energy is split between shrinking agents to fit on-device and watching what they do once they're out of the sandbox.

📚 Read — 4 items

📚 LycheeMemory V2: Efficient Long-Term Memory for LLM Agents via Semantic Segment-Level Consolidation — LycheeMemory V2 batches agent interactions into semantic segments for consolidation — a concrete approach to long-term memory that avoids unbounded context growth.

📚 Research Assistant: AstraZeneca's Agentic System for R&D — AstraZeneca's paper details its production agentic system for R&D — architecture, evaluation, and operational lessons from a large org, not a toy demo.

📚 Better, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent Cost — Finds programmatic skills (code) reduce agent cost more than other skill-learning approaches — evidence for encoding reusable capabilities as code rather than prompts.

📚 Backtrader-Bench: Benchmarking LLM Agents on Algorithmic Trading with Self-Generated MCQs — Benchmarks LLM agents on algorithmic trading using self-generated MCQs — a new eval for financial reasoning if you're testing agents in trading domains.

Memory consolidation and cost reduction dominate today's reads — the field is moving from capability demos to operational efficiency.

⚙️ Ship — 2 items

📦 Anthropic SDK 0.122.0 — Adds `output_behavior` to dream creation (new memory store or update in place); Bedrock async clients now sign SigV4 off the event loop, and beta.messages.parse/stream/tool_runner are exposed.

📦 OpenAI SDK 3.1.0 — Adds WebSocket stream IDs and workload identity access token events; deprecates Sora video APIs. Ultrafast tier, structured MCP and websocket errors, and separate websocket events land. Worth updating if you use realtime or MCP over websockets.

SDK release day: both major providers shipped operator-facing changes — memory-store output behavior on Anthropic, structured MCP/websocket errors on OpenAI. Update if you run Bedrock async or realtime.

📚 Read — 2 items

📚 How we rebuilt the Auggie CLI harness make tasks 53% cheaper than Claude Code — Augment details the Auggie CLI harness rebuild, claiming 53% lower cost per task than Claude Code — the architecture decisions behind the cut are the read.

📚 InSight-doc: Agentic Visual Perception for Long-Document Understanding — InSight-doc adaptively allocates visual resolution during reasoning — better long-document understanding with lower latency and fewer hallucinations.

Cost-per-task is becoming the battleground metric for coding harnesses, while document agents get adaptive perception. Efficiency is the through-line.

🚨 Watch — 1 item

🚨 21,000 MCP servers exposed: the protocol reaches a security inflection point — 21,000 internet-facing MCP servers found, 92% lacking OAuth. OWASP's MCP Top 10 is a start, but it's reactive — the protocol's default posture is unauthenticated. Anyone running an MCP server should audit exposure now.

MCP servers are shipping without auth by default. If you've deployed one, assume it's exposed until you've verified otherwise.

⚙️ Ship — 3 items

📦 Openrouter OpenAI: gpt-oss-20b (free) now live on — OpenAI's open-weight 21B gpt-oss-20b is now free on OpenRouter — Apache 2.0, runnable locally for agent workloads.

📦 Vercel AI SDK 2.0.0 — Vercel AI SDK 2.0.0 — major version bump; the feed's release notes are empty, so check the migration guide before upgrading any agent that uses the SDK.

📦 Litellm 17 new model(s) priced in the registry — LiteLLM registry adds pricing for 17 models including DeepSeek V4 Flash/Pro and GLM 5.1/5.2 — useful for cost planning if you route through LiteLLM.

Open-weight models keep getting cheaper to run locally; the 20B class is becoming the default for self-hosted agent workloads.

💬 Community — 2 items

💬 Yadda 3.0.0: BDD in the Age of AI Agents — Yadda 3.0.0 brings BDD-style abstraction to AI agent tests — decouples what a test asserts from how the agent does it. HN thread debates whether that layer survives agent nondeterminism.

💬 Claude Code flips auto-accept on by default — Claude Code now auto-accepts tool calls by default — no confirmation before executing. Worth checking your config if you run it against anything you don't want modified.

Two config-level decisions surfaced today: how you test nondeterministic agents, and whether your coding harness asks before it acts. Both are worth making explicit.

📚 Read — 3 items

📚 DreamX-Phi 1.0 — A paper on an action-conditioned video world model for robotic manipulation using geometric attention and depth estimation.

📚 Reference-Free Post-Training of Open Large Language Models for Multilingual Machine Translation — A paper proposing reference-free quality rewards to improve open multilingual translation models.

📚 LiveAnimate — A paper introducing a 14B-parameter video diffusion transformer for real-time, long-form pose-driven human animation.

⚙️ Ship — 3 items

📦 Hermes Agent 2026.8.16 — Hermes Agent v0.20.2 rolls up ~397 PRs: multi-gateway Connections registry, MCP health checks, persisted model routes, and prompt caching for LiteLLM Claude on the OpenAI wire. Worth a test pass if you run Hermes in Docker or hosted.

📦 Z.ai GLM 5.2 (free) on OpenRouter — Z.ai GLM 5.2 is now free on OpenRouter — zero-cost access for agent builders evaluating GLM's tool-calling without API spend.

📦 LiteLLM registry: 15 new models priced — LiteLLM registry adds pricing for 15 models, including claude-mythos-5, claude-mythos-preview, gemini-robotics-er-2-streaming-preview, and gpt-live-transcribe — useful for cost tracking if you route through LiteLLM.

Mostly incremental today: a big Hermes rollup touching gateway routing and auth resolution, free GLM 5.2 on OpenRouter, and new LiteLLM pricing. Worth a test pass on Hermes if you run it in Docker or hosted.

📚 Read — 1 item

📚 AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design — AutoDesign uses a meta-harness optimizer to recursively improve a code agent for structured media generation, achieving state-of-the-art results on long-horizon design tasks.

Self-optimizing harnesses are the pattern to watch — if meta-optimization works for design agents, it generalizes to any long-horizon agentic workflow.

💬 Community — 4 items

💬 PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives — PlayWorld puts multi-modal agents into interactive video worlds with long-horizon objectives, scoring geometry consistency and goal completion — a new eval for agent-native world models.

💬 AVA-Encoder: Towards Agent-Native Video Representation Learning — AVA-Encoder uses agentic auto-encoding with knowledge graphs to learn structured video representations, aimed at cinematic video generation from human demonstrations.

💬 Claude Is Down — HN thread flags Claude outage before status.claude.com updated — 29 points, 37 comments. Operators relying on Claude for agents should treat the status page as lagging.

💬 AI-Assisted GPU Porting of a 250k Line Legacy Weather Simulation Code — HN discussion on AI-assisted porting of a 250k-line legacy weather simulation to GPU — 54 points, with real-world Fortran/nuclear simulator anecdotes in comments.

Community energy today is split between outage triage and applying agents to gnarly legacy code — reliability and real-world porting are the live debates.

⚙️ Ship — 3 items

📦 LiteLLM registry: 20 new models priced — LiteLLM registry adds pricing for 20 Azure AI models — DeepSeek V3.2/V4-Pro and GLM 5/5.1/5.2 among them. Cost tracking for those endpoints now works out of the box.

📦 Qwen3.8 27B live on OpenRouter — Qwen3.8 27B now on OpenRouter — a mid-size Qwen model for cost-sensitive agent deployments. Pricing and provider availability on the model page.

📦 Qwen Code v0.21.13: SWE-bench 500, Terminal-Bench 89 — Qwen Code v0.21.13 validated end-to-end: 500 on SWE-bench Verified, 89 on Terminal-Bench 2.0. The release is pinned to these results — a concrete benchmark point for harness evaluation.

Model availability and benchmark validation day — new Qwen options on OpenRouter and LiteLLM pricing, plus a concrete SWE-bench number for Qwen Code. If you're cost-optimizing or evaluating harnesses, today's data points are worth a look.