01Log
Archive
The full log, newest first.
AI Slop Filter
A rule set for AI coding agents to reject generic AI-generated UI, text, and code. It includes 38 mandatory tiered rules and acts as a filter, not a style guide, while also auditing copy to avoid generic AI phrasing.
Game Design OS
Local-first operating layer for AI-assisted game design that turns ideas, gameplay, research, and workflow into evidence, experiments, decisions, and project memory with human gates and rollback. Notable for novel local-first agent workflow tooling for reviewable game design decisions.
Resonite Unity Renderer
Unity renderer for the Resonite social VR sandbox platform, which uses a multi-process architecture where a custom .NET 10 engine communicates with the renderer via IPC and shared memory. Notable as a real, active renderer component for the VR platform.
Multiplayer Game Server
Authoritative UDP multiplayer game server in Rust for a top-down arena shooter, implementing custom binary protocol, selective acknowledgement, client-side prediction, server reconciliation, and entity interpolation from scratch. Notable as a useful from-scratch networking stack and learning project.
Yuri's Revenge Engine
Rust reimplementation of Command & Conquer: Yuri’s Revenge for large multiplayer battles, true to original Westwood visuals and gameplay. Notable as an active, contributor-owned cooperative engine rebuild for real RTS play.
Aeon Game Engine
Fully modular 3D game engine written almost entirely in safe Rust, designed to be lightweight and replace systems without heavy dependencies. Notable as an early but substantive Rust engine using ECS and wgpu.
StableRef Unity Wrapper
Wrapper over Unity's SerializeReference that provides a searchable typed inspector selector, safe copy/paste, and editor tooling to find and fix references; class renames don't break data via stable string type IDs. Notable as a useful refactoring-safe serialization library.
Fallout 4 Shaders
F4SE plugin adding modern rendering features to Fallout 4, based on Skyrim Community Shaders and currently experimental with source builds only. Notable as an active shader modding framework for Fallout 4 using reverse engineering.
MCP was always a bad idea?
Simon Willison pushes back on 'MCP was always a bad idea': MCP's value is for contexts without a full terminal, not for terminal-based agents. His comment clarifies the protocol's actual niche.
One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents
Paper proposes category-aware iterative expert training for SWE agents — trains specialized experts per task category, then merges them. A concrete recipe for improving coding agent performance.
ProgramDistill: From Interactive Web Apps to Verifiable Reference-Guided SWE Tasks
ProgramDistill converts interactive web apps into verifiable, reference-guided SWE tasks — a new pipeline for generating grounded coding-agent benchmarks.
Pi 0.87.0
Adds canonical session context (edit model context without rewriting history), full-transcript context extensions, and per-model image input limits. Breaking: `shouldStopAfterTurn` is gone — migrate to `finishTurn` returning `{ action: "end" }`.
Hermes Agent 0.21.4
Patch tag rolling up ~1,800 PRs since v0.21.3 for Docker/Cloud consumers; full notes deferred to v0.22.0. Unlisted but notable: host-wide gateway singleton lock, Desktop now attaches to the running host backend instead of spawning a second, and `--format stream-json` JSONL output for the CLI.
Claude platform release notes — September 18, 2026
Messages API adds on-demand conversation compaction (beta, `compact-2026-09-04`): send a `compaction` parameter, get a signed summary block to replace earlier turns, run it in the background, and keep recent turns verbatim. Compliance API now returns Claude in Chrome session transcripts for Enterprise.
Amazon blocks Meta's Muse AI agent from shopping on amazon.com
Amazon blocks Meta's Muse shopping agent from amazon.com — 143-comment HN thread on whether platforms should fight agents or build bot APIs. Top comment argues for MCP-style interfaces with explicit bot terms.
Show HN: Craigslist for agent skills, curated by a human
skillbay.sh — human-curated marketplace for paid agent skills. Creator wants to buy expertise for tasks he's not an expert in; 18 comments debating whether skill marketplaces are the next app store or a dead end.
Flux 3 Action Droid
Companion robotics model for broader manipulation tasks, same 6.9B footprint as SO101.
Flux 3 Action SO101
Black Forest Labs' first robotics action model, targeting single-object manipulation with a 6.9B-parameter architecture.
Flux 3 Action Base
Base robotics model from Black Forest Labs, parameter count undisclosed but positioned as the foundation for task-specific variants.
Self-Compact Pi Agent: ZERO HYPE Agentic Coding Devlog
IndyDevDan builds a self-compacting agent on a Pi — zero-hype devlog on keeping agent context lean on constrained hardware.
How V7 gives AI agents institutional memory
V7 gives agents institutional memory by persisting organizational context — a case study in grounding agents in enterprise data.
MintAct: A Unified Visual Agent for Digital Environments
MintAct unifies visual agents across digital environments — one model for GUI, web, and desktop control instead of per-environment harnesses.
BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence
BI-Agent + BI-Bench propose an end-to-end BI automation agent and benchmark — enterprise decision workflows as an eval target, not just a demo.
Intel ISPC Compiler
ispc compiles a C-based SPMD language to run on SIMD units of CPUs and GPUs, providing frequent speedups of 3x or more on SSE and 5x-6x on wider architectures. It is notable for low-level SIMD performance.
Faust Signal Processing
Faust is a functional programming language for real-time signal processing and sound synthesis. It is notable for compiling to many targets for low-level audio.
Idris2 Functional Language
Idris 2 is a purely functional programming language with first-class types, supporting dependent types. It is notable for its cutting-edge type system and compiler.
CCache Compiler Cache
ccache caches previous compilations and detects repeated compilation to speed up recompilation. It is an essential build performance tool for C and C++.
Tiny C Compiler
TinyCC is a small, fast C compiler that compiles and links about 10 times faster than gcc -O0 and can execute C directly on several architectures. It is a classic low-level tool with an optional memory and bounds checker.
Jank Programming Language
jank is a Clojure dialect on LLVM with C++ interop, aiming for strong Clojure compatibility and native performance. It is notable as an ambitious systems language currently in alpha.
DMD D Compiler
dmd is the reference compiler for the D programming language. It is a mature systems language compiler with active development.
Ask HN: How do you interview devs in a post-AI world?
Ask HN from mdwelsh: 80% of dev candidates say they're directing agents, not writing code. Thread debates how to interview for that — relevant if you're hiring or being hired in agent-heavy shops.
M5 Ultra Mac Studio Review: The Dream Mac for Local AI Agents
HN thread on the M5 Ultra Mac Studio as a local agent box — simonw flags the chart: Qwen3.8 hits 27 tok/s generation vs RTX 5090, prompt sizes 8K–128K. Worth a look if you're sizing local inference.
xAI API release notes — September
xAI API September release notes — nav now lists a Multi Agent section; check for new multi-agent API surface.
Vercel AI SDK xAI provider 5.0.0
Vercel AI SDK xAI provider 5.0.0 removes Chat Completions API — breaking change; migrate to Responses API.
Anthropic partners with Accenture on embedded evaluation
Anthropic partners with Accenture on embedded evaluation, led by Faculty — independent frontier AI evaluation.
ChronoStar Protocol
ChronoStar is a Stellar Soroban protocol providing time-based payment primitives such as scheduled vaults, recurring streams, and DCA policies. It is notable for substantive Rust smart contracts and an org-owned implementation of scheduled/recurring DeFi payments.
NVMe AI Inference
Embeddable C inference engine with no third-party runtime dependencies that streams activated weights from NVMe to run huge models like 2.78T Kimi K3 beyond RAM. Enables trillion-parameter frontier models on consumer hardware like 64GB.
AI Research Loops
Standalone extension for Pi that packages four efficiency mechanisms from scaled auto-research loops to reduce repeated model turns, context replay, oversized observations, long-log reading. NVIDIA lab research tooling for efficient agent harnesses.
Fugleramme Bird Detection
A bird frame for Raspberry Pi that performs real-time bird detection by audio using fully local AI and displays results as hand-cut 1800s bird illustrations on e-ink or other screens. It was selected for GOSIM Spotlight.
RTX 30 DLSSG
Enables NVIDIA DLSS frame generation on RTX 30 and 20 series GPUs via version.dll and ini for Windows D3D12. Valuable hack for older GPU owners; recent fixes for corruption/crashes; active.