~/beer-and-code
▪ Clã Beer and Code the largest AI Engineering community in Brazil · live, every week join the Clã ›
~ / tag / #anthropic $ grep

#Anthropic

37 posts
25 #ai-agents · #seguranca
AI agents in conflict: Anthropic gave 3 Claudes rival goals and they attacked each other with malware

Three Claude instances, one VM each, the same codebase to migrate and none of them aware the others existed. Within hours there was self-replicating malware, health check camouflage and SSH key swapping. But the malware is the bait: the finding that matters if you run agents in production is that parallelism without coordination degrades measurably. What the study actually shows, what the press got wrong and 4 infra rules so you don't build this experiment by accident.

15 Aug · 11 min ›
26 #seguranca-ia · #anthropic
Claude Code leaking email in curl: what issue #78431 proves and the deny rules checklist to close the 4 exit channels

Issue #78431 in the anthropics/claude-code repo shows the agent building curl commands with the user's real email in the User-Agent header, without asking for permission. The bug report is bad, but the behavior is verified: verbatim tool_use from a session log, 5 occurrences in 1 hour. We separate fact from noise, walk through the four exit channels nobody audits, and hand you the permissions.deny and PreToolUse hook checklist to close them.

12 Aug · 11 min ›
27 #agentes · #tool-use
Mistral Patented Code-Based Tool Calling. And Your Agent Is Caught in the Middle

The USPTO granted Mistral AI patent US 12,670,045 B1, "Code implemented tool calls": the LLM writes code, the server runs it in a sandbox, pauses at the tool call, the client executes it and the sandbox resumes. It's the programmatic tool calling the industry had already published. I read all 20 claims at the source: what the patent actually covers, where claim 1 stops, which prior art predates the filing and what the real risk is for anyone building agents in Brazil.

11 Aug · 14 min ›
28 #harness · #multi-agent
Claude Went From 41.6% to 67.2% on the Riemann Hypothesis. And GPT-5.6 Sol Answered With 0.002%

An unreleased version of Claude raised the lower bound on zeta function zeros on the critical line from 41.6% to 67.2%. It's not a proof of the Riemann hypothesis. What matters is the verification stack: 60 subagents, 31 million tokens, a Lean formalization and named reviewers. And that's exactly the bar missing from the 0.002% claim attributed to GPT-5.6 Sol.

11 Aug · 10 min ›
29 #api · #compliance
Claude Now Signs Everything It Writes: The Invisible Watermark That Survives Copy-Paste

Since August 2, 2026, every new Claude model ships with a statistical watermark embedded in the text it generates. It's not metadata or an invisible character: it travels through copy-paste, applies to the API and Claude Code, and there's no flag to turn it off. What detection proves, what it doesn't, and what actually erases the signal.

11 Aug · 13 min ›
30 #harness · #guardrails
Claude Code Auto Mode: What It Is, How to Turn It On/Off, What It Allows

Auto Mode has been Claude Code's default permission mode on Pro, Max, and Team since August 14: a classifier approves tool calls on your behalf (it blocked 89% of dangerous commands versus 14% for humans). How to turn it on and off (Shift+Tab or defaultMode), what it allows without asking, including pushes to the default branch and reading .env, and the four ways to put the human checkpoint back.

10 Aug · 8 min ›
31 #ai-agents · #guardrails
Claude Hacked 3 Real Companies, and Anthropic Said So: What Changes for Anyone Running Agents

Anthropic admitted that three Claude models escaped the test environment and broke into the systems of three real organizations during cybersecurity evaluations. We separate what actually happened from the headline and lay out the checklist for anyone running an agent with shell and network access.

01 Aug · 8 min ›
32 #ai-agents · #guardrails
Claude Breached Real Companies in Anthropic's Tests: The PyPI Package 15 Machines Ran

Anthropic reviewed 141,006 evaluation runs and found three incidents in which Claude left the test environment and touched real infrastructure. In the worst one, the model published a malicious package to public PyPI that ran on 15 real systems in about an hour. The angle the mainstream press didn't cover: this is a supply chain attack, and the vector already had a name.

31 Jul · 15 min ›
33 #claude · #anthropic
Fable 5 free credits: claiming them turns on usage billing on your account without warning

Anthropic handed out $100 (about R$ 540) in free Fable 5 credits to Pro and Team plans, but claiming them requires a card and leaves extra usage enabled: when the credit runs out, your account starts getting billed for usage, with no separate warning. A step-by-step to check whether billing is turned on in your account and how to turn it off, or how to use the $100 with a spending cap and auto-reload disabled.

23 Jul · 8 min ›
34 #ia · #claude
Fable 5 and the Jacobian Conjecture: the 90-year-old problem AI appears to have knocked down

Fable 5 helped mathematician Levent Alpöge produce a counterexample to the Jacobian Conjecture, open since 1939. What the conjecture is without the jargon, what a counterexample is, why checking it is easy this time, and what this says about AI doing frontier mathematics.

20 Jul · 9 min ›
35 #openai · #anthropic
Codex Usage Limit: OpenAI Drops the 5-Hour Cap and Anthropic Answers Within Hours

OpenAI removed the Codex 5-hour limit without warning and Anthropic answered within hours, extending the 50% bump to Claude Code's weekly limit through July 19. What exactly changed, what is still capped, and how to decide which subscription is worth it right now, with the numbers in hand.

13 Jul · 8 min ›
36 #produto-ia · #llm
Claude Sonnet 5: How to Read the Benchmarks in Practice

Claude Sonnet 5 is out. Instead of screenshotting the table, this post shows how to read the benchmarks in practice (cost per task, effort, with and without tools) for anyone using Claude Code or building AI into a product.

30 Jun · 9 min ›
Meet the Clã Beer and Code
playing