Newly discovered PamStealer isn't your typical macOS malware
The discovery underscores the increased effort being poured into Mac infostealers.
Latest News Archive - Page 648
Supermarket chaos for air conditioner sale in France amid heatwave
Supermarket chaos erupted in France as shoppers raced to buy Lidl's discounted air conditioners and fans
llm-coding-agent 0.1a0
Release: llm-coding-agent 0.1a0 Another Fable 5 experiment. Now that my LLM library has evolved into more of an agent framework it's time to see what a simple coding agent would look like built on it. I started a new Python library using my python-lib-template-repository GitHub template repository, then ran these two prompts (here's the Claude Code for web transcript ): Write a spec.md for this project - it will depend on the latest “llm” alpha from PyPI and implement a Claude code style coding agent complete with tools for reading and editing files and executing commands Then: Commit the spec, then build it using red/green TDD in a series of sensible commits (each with passing tests and updated docs) - occasionally manually test it using the OpenAI API key in your environment Here's the spec , the resulting README file , and the sequence of commits . I've shipped a slop-alpha to PyPI, so you can run the new agent like this: uvx --prerelease=allow --with llm-coding-agent llm code It's pretty good for a first attempt! Here's the (Fable-authored) README , which lists recipes like llm code --yolo and llm code --allow "pytest*" --allow "git diff*" . It also presents a Python API based around a CodingAgent(model="gpt-5.5", root="/path", approve=True).run("Fix the failing test in tests/test_parser.py") class which I didn't ask for but I'm delighted to see implemented. Here's the suite of tools it implemented , listed using uvx ... llm tools : CodingTools_edit_file(path: str, old_string: str, new_string: str, replace_all: bool = False) -> str Replace an exact string in a file. old_string must match the file contents exactly (including whitespace) and must identify a unique location unless replace_all is true. Returns a diff of the change so it can be verified. CodingTools_execute_command(command: str, timeout: int = 120) -> str Run a shell command in the session root directory. Returns combined stdout and stderr followed by an Exit code line. timeout is in seconds (maximum 600); on timeout the whole process tree is killed. CodingTools_list_files(pattern: str = '**/*', path: str = '.') -> str List files matching a glob pattern, newest first. Skips hidden directories, node_modules, __pycache__ and (in a git repository) anything covered by .gitignore. Returns at most 200 paths relative to the searched directory. CodingTools_read_file(path: str, offset: int = 0, limit: int = 2000) -> str Read a text file, returning numbered lines like cat -n. Paths are relative to the session root. Use offset (0-based first line) and limit (max lines) to page through files too large to read in one call. CodingTools_search_files(pattern: str, path: str = '.', glob: str = None, max_results: int = 100) -> str Search file contents for a regular expression. Returns matches as path:line_number:line, capped at max_results. Use glob (e.g. "*.py") to restrict which files are searched. CodingTools_write_file(path: str, content: str) -> str Create or overwrite a file with the given content. Parent directories are created as needed. Prefer edit_file for modifying existing files. I tried it out by running llm code --yolo and then prompting: mkdir /tmp/demo and then in that folder create a simple swiftui CLI app for telling the time in ascii art Here's the transcript , in which GPT-5.5 reasoning notes that "SwiftUI isn't suitable for a true CLI" and then builds an app that outputs this on swift run AsciiTime : █ █████ ████ █ █ ███ ██ █ █ █ ██ █ ██ █ █ █ ████ ███ █ █ █ █ █ █ █ █ █ █ █ ███ ████ ████ ███ ███ █████ Tags: projects , ai , generative-ai , llm , llm-tool-use , coding-agents , claude-code , claude-mythos-fable
Nine Buddhist monks killed by child driving truck in Thailand
Nine Buddhist monks were killed after a truck driven by an 11-year-old boy struck a pilgrimage procession in Thailand.
FBI Seizes NetNut Proxy Platform, Popa Botnet
The Federal Bureau of Investigation (FBI) said today it worked with industry partners to seize hundreds of domains associated with NetNut, a sprawling residential proxy service operated by the publicly-traded Israeli company Alarum Technologies [NASDAQ: ALAR]. The action comes roughly two weeks after KrebsOnSecurity published findings from multiple security firms connecting NetNut to the Popa botnet, a collection of at least two million devices that have been compromised by malicious software with little or no consent from victims.
Microsoft launches $2.5 billion "Frontier Company" to embed 6,000 AI engineers inside enterprise clients
Microsoft is investing $2.5 billion in a new unit called "Frontier Company" that puts 6,000 engineers directly at enterprise customers. The goal is to integrate AI into core processes with measurable ROI, not more experimentation. Microsoft is positioning itself as a platform-neutral alternative to OpenAI and Anthropic, which push their own models through their own deployment companies. The article Microsoft launches $2.5 billion "Frontier Company" to embed 6,000 AI engineers inside enterprise clients appeared first on The Decoder .
After heat wave: Wildfires ravage southern France
Firefighters battle several wildfires in southern France that were fueled by strong winds. Nearly 3,000 tourists and locals were evacuated.
Trump bought Apple, Nvidia and other tech giants before tariff reversal fueled rebound
The day before President Donald Trump walked back his tariff policy turned out to be one of his busiest days of 2025 for stock buying.
Google Disrupts NetNut Residential Proxy Network Spanning 2 Million Home Devices
Google has significantly degraded NetNut, one of the biggest networks that turns home devices into rented relays for other people's traffic. Working with the FBI, Lumen, and others, Google's Threat Intelligence Group (GTIG) said this week it had reduced the network's pool of usable devices by millions. Google identifies NetNut, also tracked as Popa, as a network spread across home
Meta quietly launches vibe-coded gaming app Pocket
Meta has quietly launched Pocket, an experimental AI app that lets users generate and share interactive mini games using text prompts.
Anthropic is discussing a new custom chip with Samsung
The news comes about a week after OpenAI announced its own custom AI chip in a partnership with Broadcom.
Ransomware Groups Turn to Citrix Bleed 2, BYOVD, and Supply Chain Credentials
Threat actors associated with the Anubis ransomware operation have been observed exploiting the Citrix Bleed 2 (CVE-2025-5777) vulnerability to obtain initial access. "Although tactics differ between affiliates, common patterns emerged in tradecraft through use of legitimate Remote Management and Monitoring (RMM) tooling, credential access, and hands-on-keyboard procedures used for lateral
Amazon has deployed enough satellites to launch Leo service later this year
Leo will compete with SpaceX's Starlink, which had a four-year start over Amazon and has more than 10,000 satellites in its constellation.
What to know about the Society of St. Pius X, the schismatic group excommunicated by the Pope
The Society of St. Pius X was excommunicated by Pope Leo for ordaining bishops without his permission. Its issues with the church run far deeper.
Using DSPy to evaluate and improve Datasette Agent's SQL system prompts
Research: Using DSPy to evaluate and improve Datasette Agent's SQL system prompts One of this morning's AIE keynotes covered dspy , which reminded me I've been meaning to see if it could help me improve the system prompt used by Datasette Agent - so I fired off an asynchronous research task in Claude Code for web using Claude Fable 5: Pip install the latest Datasette alpha and datasette-agent and dspy - then figure out how to use dspy to evaluate and improve the main system prompts used by Datasette Agent for the feature where it can execute read only SQL queries to answer user questions about data. Fable chose to test using GPT 4.1 mini and nano, and identified several promising looking directions for improvements. I particularly like this one: The schema listing gives only table names; the "don't call describe_table if you already have the information" advice caused column-name guessing (page_count, o.order_id, first_name) and error-retry loops in baseline traces. Either include column names in the prompt's schema listing or soften that advice. Tags: ai , datasette , generative-ai , llms , evals , dspy , datasette-agent , claude-mythos-fable
Weber marks down grills and griddles to their best prices ever for July 4th
If our recent Decoder interview with Weber Blackstone CEO Roger Dahle has you craving freshly grilled meats or veggies, Weber just so happens to have a variety of grills, smokers, griddles, and accessories selling at big discounts ahead of the July 4th holiday. It makes a bunch of models, but we’ve landed on a range […]
‘Miracle’: Trapped man rescued eight days after Venezuela earthquakes
Attention has turned to the humanitarian response after at least 2,295 people are confirmed dead, and 13,000 homeless.
Copilot agent session streaming is now in public preview
GitHub Enterprise Cloud customers with enterprise managed users can now access GitHub Copilot agent session data across all Copilot clients, including: Cloud agents operating on github.com and data resident deployments… The post Copilot agent session streaming is now in public preview appeared first on The GitHub Blog .
Bomb blast at Damascus cafe kills nine, Syrian state media say
There was no immediate claim for the bombing at a cafe close to the Palace of Justice, a major government building.
How the North American heatwave could impact the FIFA World Cup
Which host cities and matches face the biggest risks? What is FIFA doing to protect players and how can fans stay safe?
Photos: Fans gather in the Bay Area for the World Cup match between the U.S. and Bosnia-Herzegovina
World Cup watch parties in Santa Clara and the San Francisco Bay Area brought fans together for the United States vs. Bosnia-Herzegovina match, where the U.S. earned a 2-0 win to reach the Round of 16.
Deadly blast at Damascus cafe leaves 6 dead, 22 injured
At least 6 people were killed and 22 injured after an explosive device detonated in a café in Damascus' Al-Hijaz area.
Rally for Luxembourg teacher fired over pro-Gaza posts
Students and supporters of a Luxembourg teacher fired over social media posts held a demonstration for her this week.
Anthropic reportedly explores custom chip manufacturing with Samsung while insisting Nvidia still matters
Anthropic is reportedly in talks with Samsung Electronics about manufacturing a custom AI chip. The project is still early, but Anthropic has already hired chip engineers. After OpenAI's "Jalapeño," yet another major AI company is pushing into chip development to cut infrastructure costs. The article Anthropic reportedly explores custom chip manufacturing with Samsung while insisting Nvidia still matters appeared first on The Decoder .