Developer Tools●●Solid
Lowfat – pluggable CLI filter that saved 91.8% of my LLM tokens
Claude Code hooks that cut token costs, but only matters if you're burning through CLI agent budgets.
Solve My ProblemCozy
zdkaster
156801mo ago
Back to browse
GitHub Repository
Point more. Token less. Reduce prose to its content words to cut LLM input tokens — across 20+ coding agents.
0 starsGo
Turo – An Aggressive Token-Saving Proxy for CLI AI Agents
by jjuliano·Jul 24, 2026·2 points·1 comment
AI Analysis
●●SolidBig BrainNiche Gem
Wenyan mode swaps words for Classical Chinese chars to beat CJK tokenizers.
Strengths
- •Lemma deduplication cuts tokens further than simple stopword removal.
- •Transparent fallback passes original text if compression fails.
- •Works across 20+ agents without modifying their code.
Weaknesses
- •Aggressive stripping risks losing nuance in complex reasoning prompts.
- •OpenAI tokenizers see less benefit from the Classical Chinese mode.
Category
Target Audience
Developers using CLI AI agents like Claude Code or Cursor
Similar To
TokenLimiter · LLMPruner
Similar Projects
Developer Tools●●●Banger
I built Panda to get up to 99% token savings
BERT-based filtering cuts token bills 99% without changing your workflow.
Solve My ProblemBig Brain
AssafPetronio
443mo ago
Developer Tools●●●Banger
BotBrowser – MCP server, saves 90% of tokens for web-browsing agents
Strips 90% of tokens from web pages for agents—no API key, no server, MIT open source.
Solve My ProblemWizardryDark Horse
amplifyco
114mo ago
Developer Tools●●●Banger
MCP server that strips injection vectors and cuts token costs by 93%
93% token reduction via deterministic sanitization instead of trusting raw web content.
Big BrainSolve My ProblemWizardry
timstark
204mo ago
Developer Tools●Mid
Save and store you .env vars
Yet another secret manager competing with Doppler and Infisical.
Solve My Problem
pwbgerrits
211mo ago
Developer Tools●●Solid
Token-saviour – agent skill from benchmarking 9 token-saving tools
Benchmark-backed router cuts agent token costs by 70% across four task layers.
Big BrainSolve My Problem
vagkaratzas
2415d ago