AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Code Assistant/
squeez
squeez logo

squeez

Active·★ 219·Apache-2.0·Updated 2026-10-19

squeez is an end-to-end token optimizer for AI CLIs that compresses bash output up to 95%, collapses redundant calls, and injects a terse prompt persona. It registers hooks automatically with zero new runtime dependencies, supporting multiple AI CLI hosts like Claude Code, GitHub Copilot CLI, OpenCode, Gemini CLI, and Codex CLI.

squeez sits in the Code Assistant category, so it is usually better judged by workflow fit, deployment style, and team needs than by isolated features alone. Based on the available metadata, it leans most clearly toward Bash compression: Intercepts every command via PreToolUse hook and applies smart filter, dedup, grouping, and truncation for up to 95% reduction.. A representative use case is Reduce token costs for AI CLI interactions, saving up to 92% on monthly API bills.. The listed license is Apache-2.0. GitHub stars are currently 218. Relevant tags include ai-cli, bash-hook.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool.

#ai-cli#bash-hook#claude-code#codex-cli#context-engineering#context-window#copilot-cli#gemini-cli
$ Install
$ curl -fsSL https://raw.githubusercontent.com/claudioemmanuel/squeez/main/install.sh | sh
↗ Visit site★ GitHub
01

Features

01Bash compression: Intercepts every command via PreToolUse hook and applies smart filter, dedup, grouping, and truncation for up to 95% reduction.
02Context engine: Cross-call redundancy detection using exact hash (FNV-1a) and fuzzy trigram-shingle Jaccard comparison (≥0.85 threshold).
03Summarize fallback: Outputs exceeding 500 lines are replaced with a ≤40-line dense summary; benign outputs have doubled threshold.
04Adaptive intensity: Scales compression limits based on session budget usage (Full below 80%, Ultra above).
05MCP server: Runs a JSON-RPC 2.0 server over stdio exposing 13 read-only tools for LLM interaction.
02

Why choose it

+Bash compression: Intercepts every command via PreToolUse hook and applies smart filter, dedup, grouping, and truncation for up to 95% reduction.
+Context engine: Cross-call redundancy detection using exact hash (FNV-1a) and fuzzy trigram-shingle Jaccard comparison (≥0.85 threshold).
+Reduce token costs for AI CLI interactions, saving up to 92% on monthly API bills.
+Supports 5 platforms or runtime environments.
03

Trade-offs

!The final decision should still be made through side-by-side comparison with similar tools.
04

Compatibility

Claude Code
Claude Code
Verified via docs
Copilot CLI
Copilot CLI
Verified via docs
OpenCode
OpenCode
Verified via docs
Gemini CLI
Gemini CLI
Verified via docs
Codex CLI
Codex CLI
Verified via docs
05

Quick start

1
$ curl -fsSL https://raw.githubusercontent.com/claudioemmanuel/squeez/main/install.sh | sh
06

Use cases

↳Reduce token costs for AI CLI interactions, saving up to 92% on monthly API bills.
↳Compress verbose command outputs (e.g., git logs, build errors, Docker logs) to stay within context limits.
↳Optimize multi-turn sessions by eliminating redundant information and providing structured memory summaries.
07

Alternatives

⚡Top VS:
VScode-context-engine★436→VStrueline-mcp★36→VSentroly★479→VSgortex★1.9k→
code-context-engine logo
code-context-engine★ 436
Save 94% on AI coding tokens. Index your codebase, agents search instead of reading files. Works with Claude Code, Codex, Copilot, Cursor, Gemini CLI. Local MCP server, free, open source.
VSCompare→
trueline-mcp logo
trueline-mcp★ 36
Smarter reads, safer edits. An MCP plugin that cuts token usage and catches editing mistakes before they hit disk. Supports Claude Code, Gemini CLI, GitHub Copilot, and Codex.
VSCompare→
entroly logo
entroly★ 479
The Context Engineering Engine. Your AI sees 5% of your codebase — Entroly shows it everything. 78% fewer tokens. Works with Cursor, Claude Code, Copilot, OpenClaw.
VSCompare→
gortex logo
gortex★ 1.9k
High-performance code intelligence engine that indexes repositories into an in-memory knowledge graph and exposes it via CLI, MCP Server, and web UI. Built for AI coding agents - expose only needed information - one get_editing_context call replaces 5-10 file reads, cutting token usage by ~94%.
VSCompare→
Graft logo
Graft★ 10.7k
Turbocharge Claude Code, Cursor, Codex, Gemini & every coding agent: faster, cheaper, with contextual understanding specific to your codebase.
VSCompare→
agent-shadow-brain logo
agent-shadow-brain★ 38
🧠 v6.0 Hive Mind — One open-source brain every AI coding agent can share (Claude Code, Cursor, Cline, Codex, Copilot, Windsurf). 22 novel features. Free. Local-first. MIT.
VSCompare→
agents-best-practices logo
agents-best-practices★ 2.4k
Provider-neutral Agent Skill for Codex, Claude Code, and agentic harness design.
VSCompare→
context-mode logo
context-mode★ 31.8k
Context window optimization for AI coding agents. Sandboxes tool output, 98% reduction. 12 platforms
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Compatibility05Quick start06Use cases07Alternatives
claudioemmanuel /
squeez
↗

Hook-based token compressor for 5 AI CLI hosts (Claude Code, Copilot CLI, OpenCode, Gemini CLI, Codex CLI). Up to 95% bash compression, signature-mode for code reads, cross-call dedup, MCP server, self-teaching protocol. Zero runtime deps.

219Apache-2.0
Last commit4d ago
StatusActive
CategoryCode Assistant
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.