tic-tools: 4 local-first tools for working with AI agents (Mac UI + CLI, Linux CLI) 👾
Visualizer: see what your agents are doing, live, across Claude Code, Codex, OpenCode & Hermes, no plugins needed. Real token & cost breakdowns by machine, provider, agent, harness & project, no double-counting. Subscription limits kept apart from API spend. Multi-machine, phone dashboard and Blob World for a fun view of your sessions.
Voice Hub: local speech-to-text & text-to-speech, no cloud. Pick models from the menu, use it yourself, through a macro or let agents hit it. Other machines can call it too, with a new multi-machine routing feature :)
Snippet Bar: one-click or macro snippets & small scripts.
Services Hub: see & manage the dashboards and gateways running on your Mac (wip -> multi-machine & linux).
Open source soon 👀
@AdamHoltererer this is so untrue.
i can't talk for gemini, i don't dare touch it.
gpt give your around 4k currently
claude 10k+ on opus or 4k fable
currently claude has better models and a more subsidized plan
@Ujjwal_v01@cherubim8997@theo@NousResearch@opencode hermes : opus 5.5, gpt 6.1 sol, daybreak blue, deepseek, glm and older stuff, i've been using it since a while now & love it.
open code (i don't use it anymore): glm, deepseek, kimi, gpt
and the big unlock recently for hands on work with agents was @t3dotcodes
@Ujjwal_v01@cherubim8997@theo try a different one.
by any chance, did you try Hermes by @NousResearch ?
or something like @opencode ?
cause those are some great tools.
(agents on a linux machine do less error in tool calls, if that is any help to you)
@cherubim8997@Ujjwal_v01@theo it's a *LLM & also a tool.
you can have the best tool in the world, if you don't know how to use it, what do you expect?
do you have any example of where you would want to review code yourself nowdays?
tic-tools: 4 local-first tools for working with AI agents (Mac UI + CLI, Linux CLI) 👾
Visualizer: see what your agents are doing, live, across Claude Code, Codex, OpenCode & Hermes, no plugins needed. Real token & cost breakdowns by machine, provider, agent, harness & project, no double-counting. Subscription limits kept apart from API spend. Multi-machine, phone dashboard and Blob World for a fun view of your sessions.
Voice Hub: local speech-to-text & text-to-speech, no cloud. Pick models from the menu, use it yourself, through a macro or let agents hit it. Other machines can call it too, with a new multi-machine routing feature :)
Snippet Bar: one-click or macro snippets & small scripts.
Services Hub: see & manage the dashboards and gateways running on your Mac (wip -> multi-machine & linux).
Open source soon 👀