Morning Singularity Digest - 2026-07-16

Context

🎨 The open-source Claude Design alternative.

What's new

🖥️ Local-first native desktop app for macOS and Windows.

Key details

• 🖼️ Your coding agent becomes the design engine: prototypes, landing pages, dashboards, slides, images & video — real files, HTML/PDF/PPTX/MP4 export.
• 🤖 Claude Code / Codex / Cursor / Gemini / OpenCode / Qwen & 20+ CLIs via BYOK.
• 🔥 Open Design 0.13.0 — Stay in Flow is here.
• Long design sessions used to break on every interruption — a run lost its place, a model picker made you guess, an export needed one more detour.

Results & evidence

• 🤖 Claude Code / Codex / Cursor / Gemini / OpenCode / Qwen & 20+ CLIs via BYOK.
• 🔥 Open Design 0.13.0 — Stay in Flow is here.
• 0.13.0 keeps the session alive: resume Codex / OpenCode / Pi / Open Design Cloud runs across turns, pick the right model faster, and hand off screenshot-backed PPTX / PDF without leaving the app.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

The agent harness performance optimization system.

What's new

Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Key details

• Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
• Language: English | Português (Brasil) | 简体中文 | 繁體中文 | 日本語 | 한국어 | Türkçe | Русский | Tiếng Việt | ไทย | Deutsch | Español Warning Official sources only.
• Install ECC only from verified channels: the GitHub repository github.com/affaan-m/ECC, the npm packages ecc-universal and ecc-agentshield, the GitHub App, the plugin slug ecc@ecc, and the project website ecc.tools.
• Third-party re-uploads and unofficial mirrors are not maintained or reviewed by the project and may contain malware.

Results & evidence

• 211.9K+ stars | 32.5K+ forks | 230+ contributors | 12+ language ecosystems | Cross-harness agent workflows Language / 语言 / 語言 / Dil / Язык / Ngôn ngữ / Idioma English | Português (Brasil) | 简体中文 | 繁體中文 | 日本語 | 한국어 | Türkçe | Русский | Tiếng Việt | ไทย | Deu...
• Production-ready agents, skills, hooks, rules, MCP configurations, and legacy command shims evolved over 10+ months of intensive daily use building real products.
• ECC v2.0.0 adds the public Hermes operator story on top of that reusable layer: start with the Hermes setup guide, then review the 2.0.0 release notes and cross-harness architecture.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.

What's new

arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.

Key details

• We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• To justify removing human experts from rubric execution, we compare rubric judgments from three human experts with those from a three-LLM judge panel on a sampled subset, and show that LLM-based evaluation is sufficiently consistent with human evaluation to...
• We then derive consensus-derived gold rubrics through two filters: a strict consistency filter, which keeps a rubric only if the three LLM judges unanimously agree on every report under the same query, and a distinguishability filter, which keeps a rubric o...

Results & evidence

• arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• This process retains 3,687 consistency-passed rubrics, of which 2,600 remain distinguishable and form the final set of consensus-derived gold rubrics.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

arXiv:2607.13639v1 Announce Type: cross Abstract: We introduce OvisOCR2, a 0.8B document parsing model.

What's new

On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Key details

• OvisOCR2 is designed as an end-to-end parser: given a document page image, it generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions.
• We build a data engine that combines filtered real-document annotations with synthetic pages whose rendered images and Markdown targets are derived from the same HTML source.
• The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy distillation into the 0.8B model, and model fusion.
• On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Results & evidence

• arXiv:2607.13639v1 Announce Type: cross Abstract: We introduce OvisOCR2, a 0.8B document parsing model.
• The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy distillation into the 0.8B model, and model fusion.
• On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Web and Tauri chat with persisted workspaces, plans, grouped live activity, file changes, embedded previews, context controls, and agent selection.

What's new

Cybara is a fully open-source, MIT-licensed AI agent platform built from the ground up with TypeScript and Bun.

It combines agents, tools, plugins, skills, MCP, ACP, LSP, browser and desktop automation, model routing, provider plan tracking, messaging int...

Key details

• I will need 12+ people that are interested in testing the Android app to get it onto the Play Store, feel free to DM me on X.
• The Apple iOS app is under review by Apple (as of this post).
• Self-hosted AI agent platform for real work: code, channels, browser automation, and on-chain execution.
• Cybara combines a Bun-based agent runtime with a web UI, CLI, desktop shells, mobile companion, encrypted local wallet controls, channel adapters, MCP support, and a broad tool layer.

Results & evidence

• I will need 12+ people that are interested in testing the Android app to get it onto the Play Store, feel free to DM me on X.
• # Clone git clone https://github.com/metaspartan/cybara.git cd cybara # Install dependencies bun install # Start full dev stack (backend + built UI + watch) bun run devThen open: - UI: http://localhost:4269 - API health: http://localhost:4269/api/health Upd...

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.

What's new

arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.

Key details

• We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• To justify removing human experts from rubric execution, we compare rubric judgments from three human experts with those from a three-LLM judge panel on a sampled subset, and show that LLM-based evaluation is sufficiently consistent with human evaluation to...
• We then derive consensus-derived gold rubrics through two filters: a strict consistency filter, which keeps a rubric only if the three LLM judges unanimously agree on every report under the same query, and a distinguishability filter, which keeps a rubric o...

Results & evidence

• arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• This process retains 3,687 consistency-passed rubrics, of which 2,600 remain distinguishable and form the final set of consensus-derived gold rubrics.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Web and Tauri chat with persisted workspaces, plans, grouped live activity, file changes, embedded previews, context controls, and agent selection.

What's new

Cybara is a fully open-source, MIT-licensed AI agent platform built from the ground up with TypeScript and Bun.

It combines agents, tools, plugins, skills, MCP, ACP, LSP, browser and desktop automation, model routing, provider plan tracking, messaging int...

Key details

• I will need 12+ people that are interested in testing the Android app to get it onto the Play Store, feel free to DM me on X.
• The Apple iOS app is under review by Apple (as of this post).
• Self-hosted AI agent platform for real work: code, channels, browser automation, and on-chain execution.
• Cybara combines a Bun-based agent runtime with a web UI, CLI, desktop shells, mobile companion, encrypted local wallet controls, channel adapters, MCP support, and a broad tool layer.

Results & evidence

• I will need 12+ people that are interested in testing the Android app to get it onto the Play Store, feel free to DM me on X.
• # Clone git clone https://github.com/metaspartan/cybara.git cd cybara # Install dependencies bun install # Start full dev stack (backend + built UI + watch) bun run devThen open: - UI: http://localhost:4269 - API health: http://localhost:4269/api/health Upd...

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Instead, you are programming the program.md Markdown files that provide context to the AI agents and set up your autonomous research org.

What's new

AI agents running research on single-GPU nanochat training automatically One day, frontier AI research used to be done by meat computers in between eating, sleeping, having other fun, and synchronizing once in a while using sound wave interconnect in the ri...

Key details

• Research is now entirely the domain of autonomous swarms of AI agents running across compute cluster megastructures in the skies.
• The agents claim that we are now in the 10,205th generation of the code base, in any case no one could tell if that's right or wrong as the "code" is now a self-modifying binary that has grown beyond human comprehension.
• This repo is the story of how it all began.
• The idea: give an AI agent a small but real LLM training setup and let it experiment autonomously overnight.

Results & evidence

• The agents claim that we are now in the 10,205th generation of the code base, in any case no one could tell if that's right or wrong as the "code" is now a self-modifying binary that has grown beyond human comprehension.
• It modifies the code, trains for 5 minutes, checks if the result improved, keeps or discards, and repeats.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.

What's new

arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.

Key details

• We address this problem by proposing a scalable pipeline for generating high-quality rubrics without human experts in the final loop.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• To justify removing human experts from rubric execution, we compare rubric judgments from three human experts with those from a three-LLM judge panel on a sampled subset, and show that LLM-based evaluation is sufficiently consistent with human evaluation to...
• We then derive consensus-derived gold rubrics through two filters: a strict consistency filter, which keeps a rubric only if the three LLM judges unanimously agree on every report under the same query, and a distinguishability filter, which keeps a rubric o...

Results & evidence

• arXiv:2607.12252v2 Announce Type: replace Abstract: Deep research agents are increasingly used to produce long-form financial reports, yet large-scale evaluation remains bottlenecked by the need for human experts to define and execute high-quality rubrics.
• We build a financial deep research benchmark from 104 real-world user queries and automatically synthesize 14,450 query-specific candidate rubrics from model-generated reports.
• This process retains 3,687 consistency-passed rubrics, of which 2,600 remain distinguishable and form the final set of consensus-derived gold rubrics.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

arXiv:2607.13639v1 Announce Type: cross Abstract: We introduce OvisOCR2, a 0.8B document parsing model.

What's new

On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Key details

• OvisOCR2 is designed as an end-to-end parser: given a document page image, it generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions.
• We build a data engine that combines filtered real-document annotations with synthetic pages whose rendered images and Markdown targets are derived from the same HTML source.
• The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy distillation into the 0.8B model, and model fusion.
• On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Results & evidence

• arXiv:2607.13639v1 Announce Type: cross Abstract: We introduce OvisOCR2, a 0.8B document parsing model.
• The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy distillation into the 0.8B model, and model fusion.
• On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

arXiv:2607.07836v3 Announce Type: replace Abstract: We present Infinity-Parser2, a large multimodal model that couples a controllable data-synthesis pipeline with multi-task reinforcement learning for end-to-end document parsing, addressing the persistent s...

What's new

First, we build a scalable synthesis engine, pairing a controllable rendering framework with an iterative refinement loop, and use it to construct and open-source Infinity-Doc2-5M: a 5-million-sample bilingual (Chinese/English) corpus spanning diverse docum...

Key details

• First, we build a scalable synthesis engine, pairing a controllable rendering framework with an iterative refinement loop, and use it to construct and open-source Infinity-Doc2-5M: a 5-million-sample bilingual (Chinese/English) corpus spanning diverse docum...
• Second, we introduce a verifiable, multi-task reward system that enables Joint Reinforcement Learning across eight co-trained objectives (document parsing, layout analysis, table parsing, math formula parsing, chart parsing, chemical formula parsing, docume...
• Third, we release two variants under a shared architecture: Infinity-Parser2-Flash, optimized for low-latency inference with a 3.68x throughput gain over Infinity-Parser-7B, and Infinity-Parser2-Pro, engineered for precision-critical settings.
• Infinity-Parser2-Pro reaches state-of-the-art 87.6% on olmOCR-Bench and 74.3% on ParseBench, surpassing DeepSeek-OCR-2, PaddleOCR-VL-1.5, and MinerU2.5, with strong generalization to charts, chemical formulas, and document VQA.

Results & evidence

• arXiv:2607.07836v3 Announce Type: replace Abstract: We present Infinity-Parser2, a large multimodal model that couples a controllable data-synthesis pipeline with multi-task reinforcement learning for end-to-end document parsing, addressing the persistent s...
• First, we build a scalable synthesis engine, pairing a controllable rendering framework with an iterative refinement loop, and use it to construct and open-source Infinity-Doc2-5M: a 5-million-sample bilingual (Chinese/English) corpus spanning diverse docum...
• Third, we release two variants under a shared architecture: Infinity-Parser2-Flash, optimized for low-latency inference with a 3.68x throughput gain over Infinity-Parser-7B, and Infinity-Parser2-Pro, engineered for precision-critical settings.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

For file submission/navigation questions, see Navigation and file context.

What's new

Windows users can jump to the PowerShell-first Windows install and release quickstart.

Key details

• github.com/code-yeongyu/lazycodex github.com/Yeachan-Heo/gajae-code Join the Discords: ultraworkers discord · gajae-code discord Important Claw Code is not the serious production project here.
• This repository is closer to a museum exhibit than a product pitch, a crustacean-run artifact kept alive by clawed gajaes, swept and labeled by agents, and automatically maintained according to the harnesses above.
• As already described in the project philosophy, this is not meant to be hand-operated like a normal product repo.
• It is an agent-managed exhibit: the harnesses plan, execute, verify, label, and preserve the artifact while the crabs keep the tank running.

Results & evidence

• No hard numbers surfaced in the source text; treat claims as directional until benchmarks appear.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

A collection of DESIGN.md files analysis by popular brand design systems.

What's new

DESIGN.md is a new concept introduced by Google Stitch.

Key details

• Drop one into your project and let coding agents generate a matching UI.
• Copy a DESIGN.md into your project, tell your AI agent “build me a page that looks like this,” and generate high-quality UI that stays visually consistent with the design language.
• Built with real design depth — including analyzed patterns, tokens, and rules — for high-quality UI generation, not surface-level outputs.
• DESIGN.md is a new concept introduced by Google Stitch.

Results & evidence

• No hard numbers surfaced in the source text; treat claims as directional until benchmarks appear.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Spotify isn't just fighting one type of problem either.

What's new

AI music has been flooding the internet for the past couple of years, with the number of AI-generated songs uploaded online growing each day.

Key details

• AI took over social media, with slop content polluting feeds at every corner.
• Music streaming platforms, too, couldn’t stop this wave, as thousands of tracks of questionable quality are being uploaded each day.
• Spotify in particular has had a complicated relationship with AI music.
• On one hand, the company encourages AI-generated music on its platform, but at the same time, it has to preserve at least some quality standards.

Results & evidence

• This conflict has led to Spotify deleting over 75 million tracks from its platform, according to the company’s executive in charge of global artists, marketing, and policy, Sam Duboff.
• Duboff also noted that roughly 100,000 songs are uploaded to Spotify's servers every single day.
• The latest industry analysis says that roughly 44% of all music uploaded to streaming platforms is AI-generated.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

arXiv:2601.02023v2 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distrib...

What's new

arXiv:2601.02023v2 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distrib...

Key details

• We investigate how fact placement, corpus-level distributions, and anti-hallucination ("Don't Make It Up") prompts influence model behavior by introducing a model-agnostic extended needle-in-a-haystack benchmark designed for scalability, which we apply to e...
• Unlike prior work, we separately evaluate literal extraction, logical inference, and hallucination risk.
• We identify two critical failure modes: Distributional Collapse, where performance degrades significantly when evidence is dispersed; and a Safety Tax, where anti-hallucination prompts cause over-conservative refusal of present facts and evidence, sharply r...
• Our results suggest that many failures stem from ineffective context utilization, as models struggle to prioritize relevant information even when it is present.

Results & evidence

• arXiv:2601.02023v2 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distrib...

Limitations / unknowns

• Unlike prior work, we separately evaluate literal extraction, logical inference, and hallucination risk.
• We identify two critical failure modes: Distributional Collapse, where performance degrades significantly when evidence is dispersed; and a Safety Tax, where anti-hallucination prompts cause over-conservative refusal of present facts and evidence, sharply r...
• Our results suggest that many failures stem from ineffective context utilization, as models struggle to prioritize relevant information even when it is present.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Show HN: Rackp – a protocol for deriving fault when AI agents cause incidents

What's new

Show HN: Rackp – a protocol for deriving fault when AI agents cause incidents

Key details

• Show HN: Rackp – a protocol for deriving fault when AI agents cause incidents

Results & evidence

• No hard numbers surfaced in the source text; treat claims as directional until benchmarks appear.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.

Context

Welcome to QUATUOR.

Feel free to fork and star and be awesome.

What's new

Welcome to QUATUOR.

Feel free to fork and star and be awesome.

Key details

• I think you can make a lot of cool stuff starting from this, or from this idea.
• I can never manage to sell anything and so now I release for free & open source.

  See you around

Results & evidence

• No hard numbers surfaced in the source text; treat claims as directional until benchmarks appear.

Limitations / unknowns

• Generalization outside curated tasks is still unclear.

Next-step validation checks

• Reproduce one claim with a public baseline and fixed evaluation settings.
• Check robustness on out-of-distribution or long-context cases.
• Track whether independent teams report matching results.