jeffyaw

i was an early vagrant user, a long time tf user, and 3 year nomad/consul user. but have moved on. cf now for aws & gcloud cli in scripts for gcp. and eks/gke instead of nomad.

An AI CLI for coding - more volume per dollar, cheaper extra usage, and monthly billing instead of 5-hour reset windows. One curl installs the launcher; then run t (or typed) from any project directory.

i will be digging into the CLI option, kudos.

you're right, i remember pre 4.6 i believe 200k was the norm.

highly agree, i built typed for this exactly.

playwright is AWESOME for e2e testing (works great with electron!).

although, i use the playwright mcp. :)

it is not, it just helps to not go in the wrong direction as much. even in my comment i mentioned reviewing the code multiple times. so i do agree reviewing is essential.

minimax m3 has a 1M token context window so not sure how op is hitting this 200k. maybe the plan they're on? or some setting in some layer of whatever their dev tooling is using.

if you want to try minimax m3 with a lot of optimizations and a custom tui check out typed. and for that harness you can use it in yaw mode.

you can also use within claude code tui by running: typed cli off

i should probably change that to typed tui off/on. anyway.

try using a harness. i provide one in Yaw Mode and you can copy it and modify and use a modified if you wanted to learn and tweak yours.

i use the skills /yaw-review excessively sometimes multiple times in a row on the same pr or session. followed by most often /yaw-address-all and then /yaw-coverage to add tests and /yaw-ship-ready to make production ready.

after a few rounds of these they are not needed every time on the same codebase.

if you are desperately wishing programming to go back to the before times it will never. or it will always be there but expect to be incredibly less productive than your peers.

absolutely, for me the tui, ultracode agentic workflows, and streaming logic are far superior. the closest model is minimax 3.0 imo and i ended up adding a custom tui, agentic workflows, streaming logic and implementing skills to that (in typed) in order to get to an acceptable claude fallback. on their own i haven’t found one model comparable to claude, not even chatgpt.

i bet 10+ things you used today were authored by claude in one place or another.

february was some kind of nirvana. i do think claude code versions and what is introduced at that level is/was relevant.

but 4.8 xhigh w/ ultracode to me is just about Fable level (w/ some agents harness tweaking).

but have to switch to 4.7 xhigh and 4.6 max quite often these days.

i just use review skills on my sessions/prs extensively until they're in a good place. then coverage skills (which add test coverage).

i built in skills to work with M3 in a service called typed, an ai cli. it uses m3 under the hood (up to ~500k tokens), then switches to deepseek for up to 1M. a few bells and whistles added of typescript/python coding optimization. and just built a custom TUI frontend for it (initially works with the claude code tui and still does).

to toggle the typed tui you can run:

typed cli on

typed cli off

there is, yaw terminal.

apologies, just trying to get the word out. and why is it crap? it seems exactly relevant to what i'm replying to.

check out yaw terminal for a terminal first experience that also treats ai cli as a first class citizen. and if you're on windows is very dialed into git bash.

as an anecdote of support for yaw terminal i am currently logged in via Yaw Mode and have been continuing to use claude all day no problems while the browser is absolutely unavailable.

check out my yaw terminal. iterm2 was absolutely an inspiration.

there is a discovery layer (although, not sure if that will be the main draw), and i am vetting that discovery layer to ensure high quality of the mcp servers we have on that discovery layer.

the big value prop is only loading the mcp server or set of tools within the mcp server that is required for each prompt. this reduces wasteful context usage and clutter (and wasteful token usage).

happy to get vital-stack into the registry. when you're ready message me at support@mcp.hosting

Yaw Mode is an opinionated config overlay for Claude Code. You toggle it on and the next session inherits a curated bundle of rules, skills, and agents without touching your ~/.claude. Turn it off and you're back to vanilla Claude Code, same conversation history, nothing mutated.

The rules are mostly things I got tired of re-explaining to Claude: no speculative abstractions, validate only at system boundaries, never bypass branch protection, integration tests hit real DBs, etc. Skills and agents are bundled alongside.

Link has the full breakdown of what's inside and how the overlay keeps your conversation history safe.

Good question, the answer is simpler than the setup suggests, because for most servers mcp.hosting isn't actually in the data path.

When you install a server, your client (Claude Code, Cursor, etc.) ends up with the upstream's endpoint written into its own config and talks to it directly. No proxy, no status rewriting, x402 is whatever the upstream sends, unmodified, and your client-side payment handler runs natively.

So paid third-party MCP servers can absolutely negotiate through mcp.hosting-installed configs, and the hosted-proxy path doesn't break x402.

didn’t Anthropic block open claw usage?

it does, but i’ve found it to be buggy, got hung up when that directory was not assigned (poor ux), and just crash often. i’ve found the cli much more performant.

it's probably because they scan your codebase/directory for context reading files, searching for references, building understanding of your project. The UI versions skip that.

conversely to all the web interfaces for all major logging providers.

That's correct.

windows/linux: ctrl+shift+S

mac: cmd+shift+S

Opens the ssh connection manager. And yes, it just uses the standard ssh, psql, mysql (etc) clients. The benefit is just clicking a button to connect. This could also be done with many different strategies. bash_aliases, ssh_config. But for a nice gui and button push UX this is what is provided. Also, if connecting through yaw there is a handy Remote Sessions gui (ctrl+shift+R) to interface with tmux or screen sessions easily.

Thanks for the reply. No github link because yes, it is closed source (although, free to use). It won't change in the future as I'm trying to model it on the Sublime Text model.

I can completely understand your concern, however, many tools these days we use are closed source (warp, cursor, sublime text, termius).

I am the creator, Jefferson Hale. Here is my linkedin: https://linkedin.com/in/jeffyaw

I'm in the United States. Lake Arrowhead, CA to be precise.

bash for shell. and i am a bit biased here, but if on windows: yaw terminal