Footprints on the Road to AGI

One job: collect what has actually happened on the road to AGI, with every entry traceable to a paper, a repo, a benchmark or a video. No headline chasing and no full tool directory — what shows here is only what we read ourselves.

The shell around an agent: prompts, tool orchestration, context management and eval harnesses. Model capability only lands through the harness, so it gets its own line here.

0:26

HarnessRouter: One Interface for Every Agent Harness

@akshay_pachaar · 57 views · 2026-09-21

Agent HarnessUnified ProtocolOpen Source

SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness

As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands into long trajectories of reasoning, tool use, and feedback. SoL-Pi scales recursive auto-research across executable environments to discover reusable harness mechanisms. Four retained mechanisms improve action execution, context compaction, observation handling, and delegated reading, reducing recorded token traffic by 44.7-49.0% and API cost by about one third at comparable EdgeBench performance.

Haozhe Liu, Tian Ye, Sensen Gao·Sep 17, 2026

Agent HarnessRecursive self-improvementToken EfficiencySep 17, 2026

ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement

ModularRSI is a benchmark-disjoint, contrastive, and modular framework for generalizable agent-harness evolution. It diagnoses recurring failures from paired successful and failed trajectories, evolves five functional modules independently, and integrates validated changes into a unified harness. On Terminal-Bench 2.0 and SWE-Bench Verified, the evolved harness improves unseen in-domain and cross-domain tasks and transfers across foundation models.

Siwei Wu, Jincheng Ren, Yizhi Li·Sep 14, 2026

Agent HarnessRecursive self-improvementAI AgentSep 14, 2026

1:00

SoL-Pi: A Token-Efficient Agent Harness Built by Auto-Research Loops

@QihangCao11 · 144 views · 2026-09-11

Agent HarnessRecursive self-improvementToken Efficiency

Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agents

Language-conditioned manipulation requires both precise contact-rich control and robust reasoning over language, scenes, and long horizons. End-to-end Vision-Language-Action (VLA) models provide strong local visuomotor skills, but they are trained on in-distribution task trajectories and often fail under deployment perturbations such as semantic retargeting, goal re-binding, spatial-layout shifts, and unstable local contacts. LLM coding agents provide complementary semantic and compositional reasoning, but purely analytic primitives struggle with irregular grasping, constrained placement, and articulated-object interaction. We present Harness VLA, a memory-augmented agentic framework that exposes a frozen VLA as a retryable contact-rich primitive and composes it with a small fixed library of analytic primitives for grounding, staging, transport, navigation, and release. Rather than expanding the skill library, the harness learns the operating range of these fixed primitives from task-specific execution traces, global success rules, and failure models. By lifting semantic re-grounding, non-contact execution, and VLA re-staging to the planner while reserving the frozen VLA for local contact-rich phases, Harness VLA extends pretrained VLAs beyond their original trajectory distribution without finetuning. Across perturbed tabletop, household kitchen, and clean-to-randomized bimanual manipulation, Harness VLA improves over the strongest relevant baselines by 38.6 and 27.1 percentage points on LIBERO-Pro and RoboCasa365, respectively, and reaches 58.4% on RoboTwin C2R. Code is available at https://github.com/RLinf/RPent.

Yixian Zhang, Huanming Zhang, Feng Gao·Jul 9, 2026

VLARobot ManipulationLLM AgentJul 9, 2026

joshhhhhan/VISTAMIT

VISTA: A Visual Harness That Lets Multimodal Models Play ARC-AGI-3 From Raw Pixels

MIT's VISTA gives a general-purpose multimodal model raw PNG observations, free-form language reasoning and a lossless archive of every environment frame, then lets it play ARC-AGI-3. With Claude Opus 5.0 it completes all 25 public games and all 183 levels at a perfect RHAE of 100, using 56% fewer actions than first-time human players; with GPT-5.6 Sol it scores 98.27, all without any program synthesis.

ARC-AGI-3Embodied AIVisual Reasoning

Python1700

Two axes of AIGC: the horizontal one is model capability (image, video, 3D and world models), the vertical one is which workflow it lands in (design, office). The five topbar entries live in this row.

One admission rule: the opposition in a match is driven by live decisions, not a scripted performance. Click a card and the round starts in-page, same-origin, nothing to install. Each card states its own verifiable claim - where the decisions come from, whether the scene is rebuilt or read out of its source, and what runs on a fixed step.

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

Unreal 5.5's first-person test map, running same-origin on this site: 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