[vova_hn2]

> random LLM-generated statistic-vomit text

I do not understand why this project in particular have set you off.

Their README looks much better than many I've seen on HN:

- no annoying verbosity, that is so prevalent in AI-generated text - not too many buzzwords (they're not saying "agentic" every sentence) - it is very clear what exactly project is supposed to do and why it can be useful

Personally, I upvoted this because I wanted to do something similar for a long time but never got around to it.

[gbnwl]

Their comment reporting stats here:

“Provider: OpenAI (gpt-4o / o1)”

Uh so is it 4o or o1? Very different models. When you read this, how did you interpret this?

    “Suite: 11-task core suite (atomic coding tasks)”

- OK ill take your word for it

    “Configuration: autoroute_first=true, single_file_fast_path=false

Run Variant Token Delta (per call) Step Savings (vs Baseline) Task Success Baseline (2026-03-13) -18.62% — 11/11 Hardened A +8.07% — 11/11 Enhanced (2026-03-27) -6.73% +27.78% 11/11 Key Takeaways:

- What useful information do you glean from this vova_hn2? Perhaps Im just ignorant.

    “The ROI of Precision: While the "Enhanced" run used roughly 6.73% more tokens than the baseline per request, it required 27.78% fewer steps to reach a successful solution.”

So it actually takes MORE tokens but less “steps”? This could all use actual discussion feom the creator. A blog post or detailed comment. Instead we get this.

What sets me off is projects like this that throw random numbers and technical jargon at you because the user simply asked their LLM to do so. It gives the veneer of “oh it must be legitimate look at all the data” to people mentally stuck in 2024 not realizing anyone can generate junk and pass if off in a way that (used to be) convincing.

[verdverm]

> Personally, I upvoted this because I wanted to do something similar for a long time but never got around to it.

It's much easier to give your agents the LSP for the language(s) it's working on.

A project that turns them into MCP: https://github.com/isaacphi/mcp-language-server

[lowbloodsugar]

Easier, sure.

[verdverm]

LSP has many more "functions" that are useful to agents, such as rename and get definition/usage.

Some basic differences described here: https://news.ycombinator.com/item?id=30664671

More in depth: https://lambdaland.org/posts/2026-01-21_tree-sitter_vs_lsp/

Claude Code's instructions for LSP: https://github.com/Piebald-AI/claude-code-system-prompts/blo...

I'm not convinced that ASTs are meaningfully helpful in the grand scheme and the long-run. LSPs and code intelligence I find useful as a human and for my coding agent.