RTK

“RTK saves between 2% and 7% on costs, it highly depends of your usage”

“I just looked at the RTK gain, and it seemed to do the job.”

“RTK has save me a lot of usage there.”

“since rtk defaults to persisting that savings stat, it breaks sandboxing”

“I kept seeing the model get confused in the reasoning text and retry a command bypassing rtk”

“rtk gain mechanism is oversimplified. 1 token != 4 bytes for the standard prompt / context window at coding agent.”

“I believe RTK would work well in that use case.”

“I wrote article about it couple months ago that I didn't believe it works”

“tell your agent to use its quiet mode or rtk”

“agents on average perform *more* turns and use *more* tokens to accomplish one task”

“we're all at a loss for how effective rtk is, because there are no benchmarks measuring its performance”

“setting up a combination of both tools could reduce token consumption by 50%”

“I've watched agents go around in circles or use ridiculous workarounds after being confused by rtk output.”

“tools like RTK, Headroom, caveman mode and others do reduce input and output tokens that need to be processed”

“on my Mac I've shaved off roughly 51k input tokens, and 23k output tokens, and saved an average of 3 seconds per command”

“Sure your saved 70% tokens on that tool call, but nothing in the metrics says whether you ran 3 tool calls instead of 1.”

“forking these tools, benchmarking, verifying the output quality matches our needs”

“I just ran the steps to reproduce and got `Error: prettier produced no output` on rtk (0.42.2)”

“I doubt it's saving much, but the quality of the work feels similar.”

“although it seems an interesting idea, I am pretty sure there are many caveats”

“You can minimize the token waste using rtk as a proxy”

“for agentic coding e.g. Claude Code I found this useful”

“the model got stuck just running and rerunning and rerunning in a loop, until I killed it”

“using their default setup which is to prepend rtk to everything is more trouble than its worth”

“my agent got stuck in a loop with a faulty RTK command and could not escape it since RTK hard overwrites anything automatically”

“the number of turns to complete went up so the overall cost of the conversation was higher”

“when it doesn't support eg. a cli flag of find, it gives an error message rather than sending the full output”

“it gives an error message rather than sending the full output of the command instead”

“always uses RTK unless RTK falls apart running a tool it does not support”

“I tested both for weeks and they didn't really saved that much money for Opus model.”

“To reduce token count on command outputs you can also use RTK”

“some people see that changing how commands output stuff can confuse some models”

“tools that can help to reduce token usage cost so you can get more prompts in within your same plan”

“It uses hooks to replace certain commands with optimized ones and helps reduce the number of input tokens”

“quite useful for reducing token usage”

“seems to be a good balance between culling out of context and not just killing everything”

“In 1 week. Numbers speak for themselves.”

“That's significant token budget reclaimed.”

“Biggest surprise was rtk find — directory listings are insanely wasteful.”

“This is a "scratching my own itch" project born from using Claude-code.”