AI Coding Live News — Real-Time 𝕏/Twitter Updates & Breakthroughs

👀62

🔥96

OpenAI Solves Legacy Software Migration: MultiLangBench Lifts COBOL Pass Rate to 97.9%, Modernizing Enterprise Stacks

OpenAI officially announced breakthroughs in building and migrating software written in low-resource and legacy languages (COBOL, Fortran, Visual Basic, OCaml), introducing the MultiLangBench evaluation suite. Tackling the trillions of dollars in critical banking and enterprise mainframe infrastructure, the benchmark challenges models to synthesize a SQLite-compatible database engine across 20 distinct languages. GPT-6 Astra boosted average test pass rates from 26.8% to 87.4%, with COBOL soaring from 10.9% to 72.1%. In enterprise COBOL migration tasks, GPT-6.1 Sol achieved a 97.9% success rate while cutting inference costs by 39%.

Key Takeaways

OpenAI unveiled breakthroughs in low-resource and legacy languages with MultiLangBench, challenging models to implement database engines across 20 languages
GPT-6 Astra raised the average pass rate from 26.8% to 87.4% (clearing >83% in 19 languages), with COBOL hitting 72.1% and Fortran 83.5%
In enterprise COBOL migration benchmarks, GPT-6.1 Sol scored 97.9% (vs 5.4% for GPT-5.6 Sol) while reducing execution costs by 39%
OpenAI confirmed open-sourcing the MultiLangBench test suite on GitHub to accelerate automated modernization of mainframe software

Read full story

Linked Models & Agents:

#OpenAI#MultiLangBench#COBOL#Legacy Migration#GPT-6 Astra#Software Engineering

OpenAI@OpenAI·Collected 16d ago

👀110

🔥88

View on 𝕏 ↗

OpenAI open-sources MultiLangBench and reports low-resource language gains: GPT-6 Astra averages 87.4% building SQLite across 20 languages, GPT-6.1 Sol reaches up to 97.9% on COBOL migration

On October 19 OpenAI said its latest models are much better at low-resource languages such as COBOL, Fortran, Visual Basic and OCaml, and open-sourced MultiLangBench, which asks a model to build a SQLite-compatible database engine from scratch in each of 20 languages. GPT-6 Astra averages an 87.4% test pass rate (GPT-5.6 Sol: 26.8%), above 83% in 19 of 20 languages, with COBOL hardest at 72.1%. On a separate COBOL migration eval requiring byte-for-byte output parity after porting to C++, Java, Kotlin, Python or Rust, GPT-6.1 Sol scores 86.6% at low effort (GPT-5.6 Sol: 5.4%) and 97.9% at higher effort. All figures are OpenAI's own.

Key Takeaways

MultiLangBench: GPT-6 Astra (max) averages 87.4% across 20 languages vs 26.8% for GPT-5.6 Sol, above 83% in 19 of them (OpenAI-reported).
Per language: COBOL 10.9% to 72.1%, OCaml 25.6% to 85.8%, Fortran 21.4% to 83.5%, Visual Basic 27.5% to 89.4%.
Budget tier: GPT-6 Luna averages 40.3% at about $0.77 per sample vs $3.48 for GPT-5.6 Sol, roughly 78% cheaper.
COBOL migration: GPT-6.1 Sol scores 86.6% at low effort (GPT-5.6 Sol 5.4%) at about 61% of the API cost; 97.9% at higher effort, near saturation.
The eval setup, test suite and per-language results are on GitHub openai/multilangbench; the COBOL migration eval is not public.

Read full story

Linked Models & Agents:

⚡ GPT-6 Astra ⚡ GPT-6.1 Sol ⚡ GPT-6 Luna ⚡ GPT-5.6 Sol 🌐 OpenAI

#COBOL#Fortran#MultiLangBench#Benchmark

Dan Kornas@DanKornas·Collected 45d ago

👀41

🔥92

View on 𝕏 ↗

Dormice open-sources self-hosted E2B-compatible agent sandboxes on one machine

@DanKornas highlighted Dormice, an Apache-2.0 self-hosted agent sandbox platform: one daemon and SQLite ledger on a single machine, idempotent sandbox access by stable key, idle freeze/stop/archive lifecycle, E2B-compatible API for the official e2b SDK, plus TypeScript SDK, dor CLI, and web console.

Key Takeaways

Self-hosted on one machine with a daemon + SQLite—no Kubernetes.
Idempotent sandbox keys with freeze/stop/archive idle lifecycle.
E2B-compatible API plus TypeScript SDK, dor CLI, and web console.

Read full story

Linked Models & Agents:

#Dormice#Sandbox#E2B#Self hosted#agent infra#CLI#Apache 2.0

Dan Kornas@DanKornas·Collected 45d ago

👀50

🔥92

View on 𝕏 ↗

agmsg open-sources local SQLite messaging for CLI coding agents

@DanKornas introduced agmsg, a local messaging tool for Claude Code, Codex, Gemini CLI, and GitHub Copilot CLI. Independent agent sessions coordinate through one shared SQLite store without a daemon, network service, or human copy-paste relay; durable history can be replayed into a fresh session, with monitor, turn, both, and manual delivery modes. The Bash + sqlite3 project is MIT-licensed and offers npx, plugin, and direct-script installation paths.

Key Takeaways

A shared local SQLite transport lets independent CLI agent sessions hand work to one another directly.
Durable history can be replayed into fresh sessions, with monitor, turn, both, and manual delivery modes.
Bash + sqlite3, MIT open source, covering Claude Code, Codex, Gemini CLI, and Copilot CLI.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)

#Agmsg#CLI agents#SQLite#Claude Code#Codex#Gemini CLI#Copilot CLI#MIT#open source

Dan Kornas@DanKornas·Collected 45d ago

👀50

🔥91

View on 𝕏 ↗

Instar open-sources persistent Claude Code agents with memory and cron

@DanKornas highlighted JKHeadley/instar, an open-source framework that pairs real Claude Code processes with persistent SQLite memory, cron jobs, and messaging so agents keep working across sessions. Features include SQLite/FTS5/rolling summaries with context re-injection, priority- and quota-aware scheduled work, Telegram/WhatsApp/iMessage, per-session tools/hooks/skills/MCP, plus behavioral hooks and external-operation gates. MIT.

Key Takeaways

Real Claude Code processes plus SQLite memory so work survives across sessions.
Cron with priority/quota awareness; Telegram/WhatsApp/iMessage outside the terminal.
Per-session skills/MCP with safety gates; MIT open source.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)

#Instar#Claude Code#SQLite#Cron#MCP#Telegram#persistent agents#MIT#open source

Dan Kornas@DanKornas·Collected 45d ago

👀60

🔥90

View on 𝕏 ↗

OpenSquilla open-sources a microkernel agent with local four-tier model routing

@DanKornas highlighted TokenRhythm/opensquilla, a token-efficient microkernel AI agent for builders. SquillaRouter scores each turn locally on length, language, code, keywords, and semantic embeddings, then routes across four tiers to the cheapest capable model; the registry targets 20+ LLM backends with primary-plus-fallback. Fifteen bundled skills load on demand, and the agent can act as an MCP client or server. Local Markdown notes add SQLite full-text search and sqlite-vec semantic recall, with three sandbox policy tiers plus human-in-the-loop approvals for sensitive tools. Apache 2.0.

Key Takeaways

Local four-tier routing picks the cheapest capable model instead of always using the top tier.
Skills + MCP: 15 on-demand skills; agent can be MCP client and server.
Apache 2.0; Markdown/SQLite memory with three sandbox tiers and human approvals.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)

#OpenSquilla#token routing#MCP#Microkernel#Sqlite vec#Apache 2.0#open source

Dan Kornas@DanKornas·Collected 45d ago

👀51

🔥89

View on 𝕏 ↗

Mira open-sources a self-hosted AI code reviewer with full-repo indexing

@DanKornas highlighted miracodeai/mira, a self-hosted AI code reviewer for teams that keep diffs, indexes, review history, and CVE data in their own SQLite or Postgres. It supports OpenRouter and OpenAI-compatible endpoints including vLLM/Ollama, applies per-repo rules via .mira.yaml, and auto-reviews PRs on GitHub, GitLab, and Forgejo/Codeberg. Apache-2.0.

Key Takeaways

Self-hosted by default: code and review data stay on your infrastructure.
Full-repo indexing plus custom rules go beyond a single diff.
Supports GitHub, GitLab, and Forgejo; models via OpenRouter or local OpenAI-compatible endpoints.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)🤖 Cursor Composer (Agent Mode)

#Mira#AI code review#Self hosted#OpenRouter#Ollama#GitHub#GitLab#Apache-2.0#open source

SunsetSyntax@sunsetsyntax·Collected 46d ago

👀54

🔥90

View on 𝕏 ↗

suvadu v0.4.1: local Rust shell-history replacement with agent tracking and MCP

@sunsetsyntax highlighted AppachiTech/suvadu v0.4.1, a Rust SQLite-backed shell-history replacement that records exit code, duration, directory, and executor, and auto-detects Claude Code, Cursor, Codex, Aider and related coding agents. It exposes a local MCP server (mostly read-only tools), Prompt Explorer, and a shared skills library. Install via brew, cargo, or the install script; MIT license with about 103 GitHub stars at verification.

Key Takeaways

v0.4.1 local SQLite shell history with structured context and auto-detection for Claude Code, Cursor, Codex, and peers.
Built-in MCP plus Prompt Explorer to trace commands back to prompts and sync skills across clients.
Install via brew, cargo, or script; AppachiTech/suvadu had about 103 stars at verification under MIT.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)🤖 Cursor Composer (Agent Mode)🤖 Aider

#Suvadu#MCP#shell history#Claude Code#Codex#Cursor#open source#Rust

Dan Kornas@DanKornas·Collected 58d ago

👀60

🔥85

View on 𝕏 ↗

amux: open-source local-first control plane with atomic kanban for parallel Claude Code / Codex / Gemini CLI workers

Dan Kornas (@DanKornas) launched open-source amux: a local-first control plane for parallel coding agents as one Rust binary + SQLite. Coordinate Claude Code, Codex, Gemini CLI, OpenCode, or Ollama workers from a dashboard or phone with atomic kanban claiming, fleet visibility, origin-stamped messaging, schedules/loops, and crash recovery.

Key Takeaways

Ship: open-source local-first amux control plane (Rust binary + SQLite).
Orchestration: atomic kanban claiming so parallel workers do not collide.
Ops: fleet visibility, origin-stamped messaging, schedules/loops, model switch, crash recovery.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)🤖 openai-codex 🤖 gemini-cli

#Amux#Multi agent#Kanban#Claude Code#Codex#Gemini CLI#OpenCode#Ollama#Rust#SQLite#Open source

Dan Kornas@DanKornas·Collected 58d ago

👀38

🔥88

View on 𝕏 ↗

Sortie: Apache-2.0 orchestrator that turns GitHub/GitLab/Linear/Jira tickets into isolated coding-agent sessions via WORKFLOW.md

Dan Kornas (@DanKornas) launched Apache-2.0 open-source Sortie: define tracker, coding agent, and concurrency in WORKFLOW.md beside the repo, then schedule GitHub/GitLab/Gitea/Linear/Jira issues into isolated workspaces. Adapters cover Claude Code, Copilot, OpenCode, Codex, and Kiro, with SQLite run history, retry queues, and stall detection.

Key Takeaways

Ship: Apache-2.0 open-source sortie-ai/sortie ticket-to-agent orchestrator.
Config: single WORKFLOW.md for trackers, concurrency, and isolated workspaces.
Adapters: Claude Code / Copilot / OpenCode / Codex / Kiro with SQLite state and stall detection.

Read full story

Linked Models & Agents:

🤖 Claude Code (Anthropic)🤖 openai-codex

#Sortie#Orchestrator#WORKFLOW.md#Claude Code#Codex#Copilot#Linear#Jira#Apache-2.0#Open source

StackBlitz@stackblitz·Collected 66d ago

👀64

🔥91

View on 𝕏 ↗

StackBlitz Launches Bolt.new v2: Fullstack WebContainers with 1-Click Cloudflare & Vercel Deploy

StackBlitz announced Bolt.new v2 with upgraded WebContainers supporting full-stack Python/Node/SQLite and 1-click cloud deployments.

Key Takeaways

Zero local environment setup: develop UI, backend APIs, and DB in-browser with live reload
Intelligent diff inspection and version rollback with instant GitHub repo synchronization
Direct integration with Claude 3.7 Sonnet and GPT-5.6 Sol multi-model generation pipeline

Read full story

Linked Models & Agents:

⚡ Claude 3.7 Sonnet (Hybrid Reasoning)⚡ GPT-5.6 Sol 🤖 Cursor Composer (Agent Mode)🌐 Anthropic 🌐 OpenAI

#Bolt.new#StackBlitz#WebContainers#Full-stack Generation#Browser IDE