Skip to content
AGI PROGRESS INTELLIGENCE

Footprints on the Road to AGI

Selected for You2,535

Frontier papers

553
HALO: A Unified Vision-Language-Action Model for Embodied Multimodal Chain-of-Thought Reasoning

HALO: A Unified Vision-Language-Action Model for Embodied Multimodal Chain-of-Thought Reasoning

Vision-Language-Action (VLA) models have shown strong performance in robotic manipulation, but often struggle in long-horizon or out-of-distribution scenarios due to the lack of explicit mechanisms for multimodal reasoning and anticipating how the world will evolve under action. Recent works introduce textual chain-of-thought or visual subgoal prediction within VLA models to reason, but still fail to offer a unified human-like reasoning framework for joint textual reasoning, visual foresight, and action prediction. To this end, we propose HALO, a unified VLA model that enables embodied multimodal chain-of-thought (EM-CoT) reasoning through a sequential process of textual task reasoning, visual subgoal prediction for fine-grained guidance, and EM-CoT-augmented action prediction. We instantiate HALO with a Mixture-of-Transformers (MoT) architecture that decouples semantic reasoning, visual foresight, and action prediction into specialized experts while allowing seamless cross-expert collaboration. To enable HALO learning at scale, we introduce an automated pipeline to synthesize EM-CoT training data along with a carefully crafted training recipe. Extensive experiments demonstrate that: (1) HALO achieves superior performance in both simulated and real-world environments, surpassing baseline policy pi_0 by 34.1% on RoboTwin benchmark; (2) all proposed components of the training recipe and EM-CoT design help improve task success rate; and (3) HALO exhibits strong generalization capabilities under aggressive unseen environmental randomization with our proposed EM-CoT reasoning.

Quanxin Shou, Fangqi Zhu, Shawn Chen · 55 views · Feb 24, 2026
VLARobot Manipulation
Feb 24, 2026
Reinforcement Learning Based Multi-Terrain Adaptive Whole-Body Control for a Quadruped Manipulator
02Paper

Reinforcement Learning Based Multi-Terrain Adaptive Whole-Body Control for a Quadruped Manipulator

Legged robots equipped with manipulators offer promising capabilities for loco-manipulation tasks in unstructured environments. However, achieving task-consistent end-effector behavior during locomotion remains challenging due to terrain-induced disturbances and the strong coupling between locomotion and manipulation, particularly for small-scale quadruped platforms. A reinforcement-learning based whole-body control framework is proposed for quadruped loco-manipulation on complex terrain. The proposed framework combines a hybrid task representation with task-prioritized whole-body learning to coordinate locomotion and manipulation while preserving task-consistent end-effector behavior. A lightweight quadruped manipulator platform is further developed to validate the proposed framework on a small-scale quadruped system. Real-world experiments demonstrate successful deployment on complex terrains such as slopes up to 30 degrees, stairs, and deformable terrain. These results demonstrate the feasibility of deploying learned whole-body control policy for quadruped loco-manipulation in real-world environments and highlight the importance of task-consistent whole-body coordination for terrain-adaptive loco-manipulation.

Yanzhou Jin, Fei Suo
Dense Temporal Motion Retargeting for Legged Robots
03Paper

Dense Temporal Motion Retargeting for Legged Robots

DTMR jointly optimizes reference phase increments and PD joint targets, deforming motion only where robot dynamics require it. Across 999 clips it improves retargeting precision while running about 19–21 times faster than STMR.

Jaeryeong Kim, Taerim Yoon
GUARD: Stabilizing Recursive Self-Improvement with Regret Guidance
04Paper

GUARD: Stabilizing Recursive Self-Improvement with Regret Guidance

Autonomous AI agents pair a frozen base model with an editable harness of prompts, skills, tool rules, and memory. Recursive self-improvement (RSI) adapts this harness from experience without weight updates, yet self-editing without external graders often backfires: reflection and scalar scores conflate what an edit newly broke with what remains unsolved, causing local regressions, skill forgetting, and memory bloat. We introduce GUARD (Grader-free Unsupervised Environment Design [UED] with Adaptive Rule Distillation), a closed-loop framework for reference-free continual harness evolution. Our key insight is to decompose each task into pre-execution self-checks and guide harness updates via directional regret. Each round grades trajectories against these checks, pinning mastered rules while prioritizing newly broken behaviors, and selects among parallel candidate edits by critic-free group-relative fitness. A Soft-HURT gate then promotes an edit only when gains outweigh regressions, scoping specialized fixes to prevent over-triggering and guaranteeing non-regression on its gated buffer. At test time, one frozen harness runs in a single pass. Across benchmarks on memory, skill evolution, continual task streams, and domain reasoning, GUARD consistently improves performance without catastrophic forgetting or memory pollution while substantially cutting deployment token usage.

Minghe Zhang, Xuetao Kang

LLM

120

Foundation models: pre- and post-training, architecture and long context, inference serving and cost per token.

Harness

110

Agent runtimes: plugins and sandboxes, tool surfaces, skill systems, long-term memory and self-improvement.

AI Coding

118

Code generation, repair, refactoring, software engineering tasks

Skills

20
Developmentmattpocock/skillsFeatured

The 38 skills Matt Pocock uses every day for real engineering

Editor's takeSells parts, not a process: 38 SKILL.md behind a strict user/model-invoked contract, a SCOPE.md admission bar and a shared GLOSSARY.md; the spec-to-tickets-to-implement-spec pipeline is the most complete agentic engineering loop in this channel.

Daily-working set of the Total TypeScript author: 38 SKILL.md files (engineering 20, productivity 7, misc 4 frozen, in-progress 7) behind a user-invoked/model-invoked contract, a SCOPE.md admission bar and a shared GLOSSARY.md; installs as a Claude Code, Codex, Copilot or Gemini plugin, or an editable skills.sh copy.

Claude CodeCodexGemini CLI+2
claude plugin install mattpocock-skills@claude-plugins-official
Shellmattpocock289.9k
The 38 skills Matt Pocock uses every day for real engineering
Developmentobra/superpowersFeatured

Fifteen skills that give a coding agent a full development process

Editor's takeIts premise is that models lack process discipline, not coding ability. The single best idea to steal is 'Rulings, not stalls': decide by default, and stop for a human only when an action is irreversible, touches a security-sensitive surface, leaks outside the worktree, or the plan is already too broken to continue.

v6.4.1 (2026-09-18), MIT, 289k stars — the largest single skill repository of its kind. It chains brainstorming, spec writing, git-worktree isolation, planning, subagent execution, red-green TDD, two-way code review and verification-before-completion into one trunk workflow, and the README lists 16+ harnesses: Claude Code, Codex, Cursor, Gemini CLI, OpenCode, Kimi Code, Devin CLI and more.

Claude CodeCodexCursor
/plugin install superpowers@claude-plugins-official
Shellobra289.2k
Fifteen skills that give a coding agent a full development process
Official & Collectionsanthropics/skillsFeatured

The official repo from the format's author: 19 skills plus spec and template

Editor's takeIf you want to know how a SKILL.md is actually supposed to be written, read this instead of any second-hand tutorial. Two things to know first: docx, pdf, pptx and xlsx are source-available rather than open source (the other 15 are Apache-2.0), and the repo carries its own disclaimer — these skills are demos, so test them in your environment before trusting them with anything critical.

Anthropic's own implementation, 177k stars: 19 skills under ./skills plus ./spec (the Agent Skills specification, now at agentskills.io) and ./template as a scaffold. Its definition of a skill is folders of instructions, scripts and resources loaded dynamically, and progressive disclosure comes straight from here.

Claude Code
/plugin marketplace add anthropics/skills
Pythonanthropics177.3k
The official repo from the format's author: 19 skills plus spec and template
Design & Frontendnextlevelbuilder/ui-ux-pro-max-skillFeatured

Design knowledge as searchable local data: 79 styles, 192 palettes, 119 UX rules

Editor's takeIts sharpest self-description is that models can write CSS but lack a taste database. Ten rule categories are ordered by severity, with Accessibility and Touch & Interaction first as CRITICAL and Charts last — that ordering is the stance. The Design System Generator runs five parallel searches, and the repo states plainly: do not persist unverified output, so a generated design system has to be checked before it lands in your repo.

v2.13.0, MIT, 129k stars, site uupm.cc. Instead of lecturing the model it ships structured catalogues the agent queries before writing code: 79 UI styles, 192 palettes with reasoning, 74 type pairings, 119 UX rules, 105 icon suggestions, 17 GSAP presets, 25 chart types, 22 stacks and 34 landing-page patterns.

Any agent
npx ui-ux-pro-max-cli init --ai claude
Pythonnextlevelbuilder129.3k
Design knowledge as searchable local data: 79 styles, 192 palettes, 119 UX rules
KnowledgeGraphify-Labs/graphifyFeatured

Map the whole project into a queryable knowledge graph instead of grepping

Editor's takeThe design that matters most: every edge is labelled EXTRACTED or INFERRED, so you can tell what the source explicitly contains from what the tool inferred — which makes an agent's conclusions auditable, a hard requirement in production. It is not a vector index: no embeddings, no vector store, a real graph you can explain, path and query. In its own benchmark, building the graph costs zero LLM spend and ingest is an order of magnitude cheaper, a structural win from local parsing.

YC S26, Apache-2.0, 120k stars, Python 3.10+. /graphify . maps code, docs, PDFs, images and videos into a graph and emits a clickable graph.html, a human-readable GRAPH_REPORT.md and a machine-readable graph.json. Code goes through tree-sitter AST parsing (about 40 languages): deterministic, no LLM calls, nothing leaves the machine.

Claude CodeCodexCursor
uv tool install graphifyy && graphify install
PythonGraphify-Labs119.9k
Map the whole project into a queryable knowledge graph instead of grepping
Design & FrontendLeonxlnx/taste-skillFeatured

Taste Skill: 13 anti-slop frontend skills that turn AI tells into countable rules

Editor's takeThe lesson worth stealing is rules expressed as countable mechanical tests: eyebrow count ≤ ceil(sections/3), bento cells equal to content items, zero em-dashes on the page, palette families rotated across projects. Each one aims straight at the model repeating itself instead of stopping at an adjective. The repo is equally honest about its edges: v2 self-declares experimental and is still moving, §13 explicitly excludes dashboards, data tables, multi-step wizards, code editors and native mobile, and the only test in the repo guards the source-invocation semantics of the skill.sh shell helper rather than the design rules.

One MIT repo holding 13 skills: ten that output code and three that output reference images only, all installed into Claude Code / Codex / Cursor with a single npx skills add. The default design-taste-frontend v2 (experimental) is a 1,228-line SKILL.md: first the model must emit a one-line "design read", then it sets three dials (DESIGN_VARIANCE / MOTION_INTENSITY / VISUAL_DENSITY), then it implements against 14 chapters of hard rules - install the official package for corporate briefs, label aesthetic borrowings in comments, zero em-dashes anywhere on the page, eyebrow quota computed from section count, bento cells equal to content items, and finally a 50-plus box pre-flight check. The repo also ships research/laziness, a literature directory using controlled studies to explain why models cut output short.

Claude CodeCodexCursor
npx skills add https://github.com/Leonxlnx/taste-skill
JavaScriptLeonxlnx94.8k
Taste Skill: 13 anti-slop frontend skills that turn AI tells into countable rules

Leaderboard

519
Artificial Analysis Intelligence IndexOfficial intelligence index (higher is better)
Claude Fable 5.2 (max)64.56
Claude Fable 5.2 (xhigh)62.22
Claude Fable 5.2 (high)61.82
Gemini 4 Carbon (high)60.03
Claude Opus 5.5 (max)59.11

Source · artificialanalysis.aiofficial Intelligence Index v5synced Nov 14, 2026

Image Generation

68

Text-to-image, image editing, consistency and controllability

Audio & Video

180

Video and speech/audio: physical and temporal coherence for text/image-to-video, TTS and voice cloning, music generation and ASR.

3D & World Models

403

3D asset generation, scene reconstruction, long-horizon world prediction

Design & Frontend

88

Visual design, UI components, web and slide layout

Docs & Slides

19

Slides, Word, Excel, PDF generation and editing

Blog

143

Agent Studio

7

RetouchPi

Fifty-four professional tools, layers and masks, PSD files that open and save back - all in the browser, nothing to install.

retouchpi.com · 54 pro tools · PSD read and write · AI generate & edit

ThreeDream simulation engine

Built for games and for physical AI: an ultra-realistic, ultra-high-performance simulation environment with ultra-low-cost procedural data; Unreal-grade visuals open and play in the browser, bridging the physical world and intelligent agents.

Games meet physical AI · Unreal-grade on the web · Open and play

CodeWaifu, your caring coding desktop sprite

A caring coding sprite on your desktop and a cockpit that never stops: the work resumes itself after a crash or a reboot. Every coding terminal, ssh session and local shell in one tree, every start, finish and permission request reported in time. Open source, one-line install.

Your caring coding desktop sprite · Resumes after crash & reboot · First-rate terminal / ssh / shell

Agent digital twin office

A 100 km 3D city that opens in seconds — crawling, blogs, papers, industry and investment analysis, all watchable in one digital twin.

100km city · opens in seconds · Live task stream · Agent-maintained

Robotic arm workbench

An SO-101 6-DoF arm running MuJoCo physics in your browser: joint teleoperation, IK end-effector dragging and gripper pick-and-place into a basket. No install.

MuJoCo WASM physics · SO-101 · 6 DoF · Contacts / telemetry HUD

Joint actuator anatomy lab

Take a humanoid joint module apart layer by layer: brushless motor, magnetic encoder, planetary / harmonic / cycloidal drive and output flange. Switch architectures on one page — exploded view plus analytic kinematics.

Three gearbox types · Planetary / harmonic / cycloidal · Analytic kinematics · explode

AI Games

5

Under constructionThe games system is still under construction — everything listed starts in-page, and the library keeps growing.

ThreeDream FPS arena

ThreeDream FPS arena

A first-person match where every bot across from you is driven live by a policy this repo trained, the whole round stepping at a fixed 60 Hz: builtin, Rust/wasm or Rapier behind one physics interface, nothing scripted.

WASD move · mouse look · left-click fire · shift sprint · R reload · esc release

ThreeDream UE level

ThreeDream UE level

Unreal 5.5's first-person test map, running same-origin in the browser: the level, the materials, the lights and four weapon data assets are read out of the project's own .umap / .uasset bytes, six soldiers are driven by a four-state brain on one fixed 60 Hz step, and the sidebar lists every substitution the page had to make.

WASD move · mouse look · left-click fire · shift sprint · space jump · R reload · F pick up · esc release

ThreeDream CARLA town drive

ThreeDream CARLA town drive

The whole of CARLA Town10HD read out of its .umap and OpenDRIVE road bytes, 4,474 actors back on the project's own coordinates; 32 cars decide live over the same lane graph - car-following, braking, lane changes - and you sit in one of them, on the wheel or on autopilot.

W throttle · S brake · A D steer · space handbrake · T autopilot · C camera

ThreeDream factory town

ThreeDream factory town

The abandoned-factory town's bytes come from two places: 3,571 actors and a 509x509 landscape heightfield read out of the Unreal project's own mainmap.umap, 3,283 meshes resolved through an unlicensed content pack pinned by revision on its own mirror; six soldiers run live on a four-state patrol / engage / advance / down brain over one fixed 60 Hz step, standing at random walkable spots wave by wave and getting back up elsewhere once downed, with four guns on the ground each wave to swap for with F and a kill quota that opens the next one (?arena=0 gives the same town back without rounds); and the sidebar lists the 101 placements still missing their mesh instead of hiding the holes.

WASD move · mouse look · left-click fire · shift sprint · space jump · R reload · F pick up · esc release

ThreeDream Solis Reach

ThreeDream Solis Reach

A whole mirror-salt solar city generated at runtime: 4,570 built pieces, 3,197 hillside homes, 71 heliostats and a three-ring armillary surround one live sea; 11 salt skiffs query the same heightfield that is drawn every frame, and orbit or fly cameras let you cross the scene - this is not a pre-baked tour.

drag orbit · wheel zoom · WASD fly · E Q up/down · shift boost · mouse look

Frequently asked

What is this site?
An AGI progress digest: paper deep reads, open-source projects, demo videos and leaderboards across models, agents and runtimes, skills, multimodal and embodied AI.
Where does the content come from?
Public sources: arXiv papers, official releases and blogs, GitHub repositories, demos on X and YouTube, and evaluators' own leaderboards. Every entry keeps its source link and author credit.
Are the paper and project write-ups summaries?
No. Papers are written up from the full text as bilingual long reads with formulas, experiment tables and method diagrams; open-source projects are read through their README and source before a bilingual write-up with a hands-on verdict.
Where do the leaderboards come from?
From each evaluator's own official board, synced on a schedule, with scrape and official-update timestamps shown.
Can I cite this site?
Yes — cite the canonical URL each page declares. Third-party papers, videos and images remain the copyright of their original authors.

As an Amazon Associate, we earn from qualifying purchases.