A developer known as dv-hart has released Newspeak, a Claude Code skill that compresses agent output by stripping filler words and hedging while preserving code, diffs, and security warnings, using th…
Alessandro Perrotta, developer of the pi coding agent, reimplemented the terse-speak 'caveman mode' natively in pi using a 40-line extension and skill, after finding the upstream caveman project too b…
A new technique called Prompt+Steer, which combines prompting with activation steering, reduces output tokens by an additional ~13% beyond prompting alone on the CodeXGLUE code-to-text task, according…
Developers are creating tools to reduce token consumption by AI coding agents, addressing the rising cost of tokens in 2026. Three projects—rtk, caveman, and superpowers—take different approaches: rtk…
A developer released token-saviour, an agent skill that routes coding tasks to the most token-efficient tool based on a benchmark of nine tools, achieving up to 69.6% token reduction on comprehension …
A developer argues that focusing on token optimization for AI agents is a distraction from the real cost drivers, which are agentic loops involving multiple subagents and long-horizon execution. The d…
A developer built paraNOyar, a tool that forces coding agents to explain terminal commands before execution, enabling non-technical founders to understand and approve actions. The tool is installed vi…
Developer Steve Solun released Ctx, an open-source tool that reduces token costs by selecting only relevant skills, agents, MCP servers, and harnesses before a session begins, rather than compressing …
An evaluation of four agentic tools claiming to reduce token usage found no intervention demonstrated robust token savings across all models in realistic end-to-end tasks. The study tested caveman, rt…