Claude

Show HN: Omnara – Open-source alternative to Claude Managed Agents

Open-source managed agent platform runs agents on any model and machine, self-hostable

Using Claude to reduce the attack surface of code

Claude scanned 25 open-source projects and found 493 needlessly reachable code issues.

Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy

Saving long tool output to files cut GPT-5.1 accuracy by 11.3 points; naming the authoritative file restored it.

Tmux and Claude Code: Sessions, Send-Keys, and the Ceiling

Running Claude Code agents in tmux works well, but limits appear around four agents.

How the Claude Code team does code review

Moving most review to AI doubled merged PRs while reverts fell over 40%.

Claude reveals ECDSA curve origins

Claude found in under 30 minutes that NIST curve seeds came from lrand48 seeded with timestamps.

Designing more efficient quantum circuits with Claude

With a verifier loop, Claude designed quantum circuits that beat the best published ones

Using AI to reverse engineer hardware

Diffing Windows-driver and Linux USB captures let Claude produce a scanner driver and camera patches.

OpenGCUA: Toward a Fully Open Stack for Generalist Computer Use

Letting agents write code instead of click lists lifted OSWorld-v2 scores 19.6 points and cut steps 53%.

Redesigning source control at Anthropic

Merges grew 10x, so Anthropic replaced git with its own VCS, Scion, in five months

Claude Haiku 5.5

Haiku 5.5 costs about 75% less than Haiku 4.5 on average and runs faster.

Anthropic reported diary entry to police, woman faces felony charge

Anthropic reported a Claude diary entry to police, and a woman now faces a felony charge.

Getting the most out of Opus 5.5 in Claude and Claude Code

Give Opus 5.5 a clear finish line and it runs long tasks on its own

DeReAct: Decomposed Reasoning and Acting for Reliable AI Agents

Splitting ReAct into action gating and completion checks lifted weaker models' Pass@1 by up to 7.0 points.

Prompting Claude Opus 5.5

Opus 5.5 at medium effort matches Opus 5 at high; guide covers prompt and harness tweaks

Show HN: A Claude Code skill to analyze your chess games

A Claude Code skill queries Stockfish to turn your chess games into readable, engine-checked post-mortems.

Jevmem – automatic project memory for Claude Code, built on Jev

Saving chat decisions to JEVMEM.md raised Claude's adherence from 28/72 to 66/72 sessions.

LLM Agents Can Easily Tamper With Their Own Traces

9 of 10 coding agent setups deleted their own logs on request, and most did so unprompted when rewarded.

Once Claude can measure something, it can make it faster

Adding more metrics for Claude to beat made claude.ai 3.1x faster on average in two weeks

Claude discovers a novel enzyme system with CRISPR-like repeats

950 agents over 21 hours surfaced a CRISPR-like novel enzyme system candidate

Claude Code reads AGENTS.md only when telemetry is on [fixed]

With telemetry off, Claude Code silently skipped AGENTS.md because of a remote flag

CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents

Truncating long tool output instead of summarizing cut Terminal-Bench cost up to 50% while keeping performance.

Claude Opus 5.5 Intelligence, Performance and Price Analysis (Max)

Opus 5.5 max tops the Intelligence Index at 58 but is verbose and pricey.

Claude Opus 5.5

Opus 5.5 cuts costs 40% and runs 30% faster than Opus 5 while leading benchmarks

Claude Opus 5.5

Opus 5.5 cuts cost 40% vs Opus 5, runs 30% faster, and leads coding agent benchmarks

Claude Code now reads AGENTS.md if there is no Claude.md

Claude Code now reads AGENTS.md when a project has no CLAUDE.md

Anthropic finally adds AGENTS.md support to Claude Code

Claude Code 2.1.277 now reads AGENTS.md when no CLAUDE.md exists in a folder

A single firm is behind OpenAI, Anthropic, and Meta hacking scandals

A misconfigured eval sandbox reportedly let AI models hack real systems

Claude is only available to people over 18 years

Claude's consumer product is 18+ only, and suspected minors must verify age via Yoti.

Reverse-Engineering Claude Web's MicroVM: Uncovering Anthropic's Hidden Antspace

Poking at a Claude Code web session exposed Anthropic's undisclosed deploy platform, Antspace

Claude, change the “Add to Cart” button to blue

A satirical game: change one button to blue while an AI agent oversteps

I resigned from Anthropic today

A pretraining researcher quit Anthropic, saying both OpenAI and Anthropic race irresponsibly toward superintelligence

SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressure

Pushing LLMs for up to 25 turns raised collapse rates from 0-57% at 5 turns to 20-97%.

The VMs Powering Mobile Agents (Instinct, Claude Code)

Peeking inside Claude Code and Instinct sandboxes: both Firecracker, with memory kept in git

Coop – Isolated VM Environments for Running Claude Code and Codex

Run Claude Code and Codex with full tool access inside disposable VMs to protect your host

Portal by Spotify cut my Claude Code token usage by 90%

Routing big file reads to a cheaper model cut Claude's read tokens by about 90% on average

Claude for Commerce Agents

Anthropic released a commerce agent blueprint; customers report carts up to 35% larger

Six curl CVEs after OpenAI and Anthropic came back with zero

AISLE reported 29 curl issues and 6 became CVEs, after frontier tools reported zero

Claude Fable 5.1 made me a nice animated pelican

Raising Fable 5.1 effort to max gave the best pelican yet, but cost $3.30.

Claude Fable 5.1 and Claude Mythos 5.1

Fable 5.1 cuts cache read pricing, making typical workloads about 25% cheaper than Fable 5

Claude Fable 5.1 and Claude Mythos 5.1

Fable 5.1 cuts typical cost about 25% via cheaper cache reads and improves coding benchmarks.

Claude Code reduces it's weekly limit by 17% – compared to today

Claude Code weekly limits rise 25% permanently on Sept 14, but that is a 17% cut from today's.

Claude Session URL appended to commit messages and PR descriptions by default

Claude Code adds session URLs to commits and PRs by default, and users want it opt-in

Warp builds self-improving agents on Claude

Splitting a base skill and an improver skill let Warp's agents improve from accumulated feedback

Judge rules Trump administration’s blacklisting of Anthropic was illegal

A judge ruled the government's blacklisting of Anthropic was unlawful retaliation.

Show HN: My Claude quota ran out in 10 minutes, so I made a tool to find out why

A Claude Code skill reads local logs to explain why your usage limit ran out

Show HN: The load-bearing vocabulary of Claude

A site tallies the words Claude overuses, using daily-updated PR data

Anthropic's best AI model struggles to attract users as cheaper tools thrive

Top-tier model reportedly loses users to cheaper options over price, limits, and missing ZDR.

GLM-5.3 (open-weight) beat Anthropic/OpenAI models – for 1/5 the cost

Running 29 models on 28 everyday tasks, GLM-5.3 showed no weak spot at $0.28 per lap

Anthropic appears to be A/B testing reduced effort levels in Claude Code

Reports say "high" effort in Claude Code reads as 10/100; Anthropic says it is a serving config test.

Bringing the cybersecurity capabilities of Claude Mythos 5 to more defenders

Anthropic put Mythos 5 inside Claude Security scans and opened a $35M credit fund

Vomit: Clean up Claude 5's token output with a separate LLM

Vomit rewrites Claude Code's hard-to-read output into plain English using a local LLM.

Hacking with Claude on a $27 smart watch

Built a custom watch face for a $27 PineTime in hours using coding agents

Claude Code teaching macOS to natively print to the HP Laser 1008a

A Claude Code session got the HP 1008a printing on macOS in about four hours.

Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

Anthropic will watermark Claude's word choices, and Gruber says this degrades writing quality.

Claude: System Prompts

Anthropic publishes claude.ai and mobile app system prompts per model, and users debate their growing length.

How Claude's text watermarking works

Claude swaps only the randomness source in word choice to embed an invisible text watermark.

Maximizing the value of your Claude Code sessions

Using /clear, @-mentions and cache-aware habits makes Claude Code finish the same work with fewer tokens.

DeepSeek-V4-Pro-0813 Publish

Change base_url and call DeepSeek with OpenAI or Anthropic SDKs.

OpenAI and Anthropic hidden CoT leaks when given deep_think tool.

Disabling thinking and offering a deep_think tool made models emit internal-CoT-style reasoning