Feed aggregator

Article URL: https://changelog.qgis.org/en/version/4.0/

Comments URL: https://news.ycombinator.com/item?id=47285846

Points: 26

# Comments: 0

A follow up to this thread where the CEO of fivetran asked anthropic to build group chat.

https://news.ycombinator.com/item?id=47280200&seen=177285783...

I don't know. I don't think this is novel or genuinely interesting but obviously there is a need and I'm sure there are lots of these chats with AI now available for teams. I got a Claude Max subscription recently for the open source project Go Micro so what better use than showcasing how you can use the framework for stuff like this...

Comments URL: https://news.ycombinator.com/item?id=47285734

Points: 1

# Comments: 0

I've been building Voie — a directory of European infrastructure providers for developers who want to run on European rails.

One thing that kept coming up: people wanted to earn by recommending European tools they already use. So I mapped which providers have affiliate/partner programs and linked directly to their application pages.

No paid placement. No middleman. Just the programs that exist.

Comments URL: https://news.ycombinator.com/item?id=47285731

Points: 1

# Comments: 0

Article URL: https://github.com/ChameleonTeaming/rabbithole

Comments URL: https://news.ycombinator.com/item?id=47285713

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=cqHH2NXcf5E

Comments URL: https://news.ycombinator.com/item?id=47285684

Points: 1

# Comments: 0

Article URL: https://www.braindance.dance/

Comments URL: https://news.ycombinator.com/item?id=47285657

Points: 4

# Comments: 0

Article URL: https://github.com/bugrax/git-lanes

Comments URL: https://news.ycombinator.com/item?id=47285631

Points: 3

# Comments: 2

Single Rust binary (~3 MB) that reads the OS accessibility tree and gives every UI element a REST endpoint. Click buttons, type text, toggle checkboxes — all via JSON. Works as an MCP server too, so Claude/Cursor/Windsurf can control any desktop app out of the box.

Windows + Linux + macOS. MIT licensed.

Comments URL: https://news.ycombinator.com/item?id=47285616

Points: 2

# Comments: 0

so i’ve been experimenting with a small AI-based project recently and started thinking about logging around prompts, responses, and model calls etc etc.

for traditional systems observability tools handle most of this, but with LLM-based apps it feels less clear what the standard approach is, especially if you need proper audit trails for debugging or compliance.

curious how teams are handling this in production

are people mostly building their own logging pipelines, or are there reliable tools/platforms that help with storing and auditing LLM interactions?

Comments URL: https://news.ycombinator.com/item?id=47285609

Points: 2

# Comments: 0

Article URL: https://code.claude.com/docs/en/scheduled-tasks

Comments URL: https://news.ycombinator.com/item?id=47285607

Points: 1

# Comments: 0

Article URL: https://www.technologyreview.com/2026/03/06/1134012/is-the-pentagon-allowed-to-surveil-americans-with-ai/

Comments URL: https://news.ycombinator.com/item?id=47285572

Points: 3

# Comments: 0

Today I learned that they ship the entire CLI with the agent SDK. It's all there...

Do we think this was a mistake? ----

node_modules/@anthropic-ai/claude-agent-sdk/cli.js.

It's the bundled, minified CLI binary for Claude Code itself (the tool you're talking to right now). Key details: - Version: 2.1.71, built on 2026-03-06 - ~13,800 lines of heavily minified/bundled JavaScript - Shebang: #!/usr/bin/env node — it's meant to be run as a CLI executable - Copyright: Anthropic PBC, with a cheeky note: "Want to see the unminified source? We're hiring!" - What it contains: The entire Claude Code CLI application bundled into a single file, including: - Onboarding/setup screens - Policy/managed settings loading - Debugging/inspector detection - UI rendering (using Ink/React for terminal UI) - Prefetching logic - Error handling and exit codes - The main entry point (main function) that starts the interactive Claude Code session Essentially, when you run claude in your terminal, this is the file that executes. The @anthropic-ai/claude-agent-sdk package ships it so you can programmatically invoke Claude Code's capabilities — the SDK wraps this CLI as its underlying engine. --- Yes, the agent loop is in there. Here's what I can see from the minified code:

1. i6z function — This is the in-process agent runner, which logs "Starting agent loop for ${agentId}". It sets up agent identity (agentId, parentSessionId, agentName, teamName, color, planMode) and runs the loop. 2. l6z function — A poll loop for in-process teammate agents. It continuously checks for: - Pending user messages - Mailbox messages from other agents - Shutdown requests - New tasks from the task list 3. Main loop model orchestration — There's extensive logic around mainLoopModel that handles: - Model selection based on permission mode (plan mode uses Opus, etc.) - System prompt assembly (tools, MCP instructions, skills, memory, changed files) - Token usage tracking and budget management - Compaction (context window management when conversations get long) 4. Multi-agent coordination — The loop supports a team-based agent architecture with messaging between agents (agentName, teamName, teamLead, mailboxes, shutdown requests). So yes — the full agentic loop (prompt → LLM call → tool execution → repeat) is bundled in this file, just heavily minified. It's the same loop powering this conversation right now. --- Full system prompts are right there in the bundle as plain text strings. Here's what's embedded:

Three identity variants: - F78 — CLI mode: "You are Claude Code, Anthropic's official CLI for Claude." - EtA — SDK mode: "You are Claude Code, Anthropic's official CLI for Claude, running within the Claude Agent SDK." - LtA — Agent mode: "You are a Claude agent, built on Anthropic's Claude Agent SDK." Full system prompt sections (line ~6213-6217): The ULz function assembles the complete system prompt, including all the sections you'd recognize: - "# Doing tasks" — instructions about software engineering, not proposing changes to unread code, avoiding over-engineering, OWASP security, etc. - Over-engineering rules — "Don't add features beyond what was asked", "Don't add error handling for scenarios that can't happen", "Don't create helpers for one-time operations" - Help/feedback info — links to GitHub issues - Tool usage instructions — Read vs cat, Edit vs sed, etc. Output style variants (line ~6517): Built-in output styles like Explanatory and Learning with their own prompt overlays. Also present: - The classifier/safety prompt (~line 2844) for security-relevant harm prevention - Subagent instructions like "You ARE the fork. Do NOT spawn sub-agents" - Git commit/PR workflow prompts - Memory system instructions

Comments URL: https://news.ycombinator.com/item?id=47285571

Points: 1

# Comments: 0

ran this over the weekend. stack was Llama 3.2 3B running locally + Keiro Research API for retrieval.

85.0% on 4,326 questions. where that lands:

ROMA (357B): 93.9% OpenDeepSearch (671B): 88.3% Sonar Pro: 85.8% Llama 3.2 3B + Keiro: 85.0%

the systems ahead of us are running models 100-200x larger. that's why they're ahead. not better retrieval, not better prompting — just way more parameters.

the interesting part is how small the gap is despite that. 3 points behind a 671B model. 0.8 behind Sonar Pro. at some point you have to ask what you're actually buying with all that compute for this class of task.

Want to know how low the reader model can go before it starts mattering. in this setup it clearly wasn't the limiting factor and also if smaller models with web enabled will perform as good( if not better) as larger models for a lot of non coding tasks

Full benchmark script + results --> https://github.com/h-a-r-s-h-s-r-a-h/benchmark

Keiro research -- https://www.keirolabs.cloud/docs/api-reference/research

Comments URL: https://news.ycombinator.com/item?id=47285569

Points: 5

# Comments: 1

Article URL: https://store.steampowered.com/news/group/4145017/view/528746884222682052

Comments URL: https://news.ycombinator.com/item?id=47285532

Points: 3

# Comments: 1

Article URL: https://www.wsj.com/business/retail/fast-free-shipping-retailers-fedex-ups-amazon-e6ba5967

Comments URL: https://news.ycombinator.com/item?id=47285511

Points: 2

# Comments: 1

Article URL: https://developercommunity.visualstudio.com/t/Comprehensive-Vim-Motions/11051331

Comments URL: https://news.ycombinator.com/item?id=47285499

Points: 1

# Comments: 0

Article URL: https://surgehq.ai/blog/enterprisebench-corecraft

Comments URL: https://news.ycombinator.com/item?id=47285498

Points: 1

# Comments: 0

Article URL: https://www.vercel-status.com/incidents/963t8qqbk77b

Comments URL: https://news.ycombinator.com/item?id=47285496

Points: 1

# Comments: 1

Article URL: https://www.infoq.com/news/2026/03/agents-context-file-value-review/

Comments URL: https://news.ycombinator.com/item?id=47285495

Points: 1

# Comments: 1

Article URL: https://www.reuters.com/legal/government/xai-loses-bid-halt-california-ai-data-disclosure-law-2026-03-05/

Comments URL: https://news.ycombinator.com/item?id=47285465

Points: 4

# Comments: 1