Le descrizioni degli strumenti sono in inglese.

Efficienza dei token

Strumenti per ridurre i costi dei token e comprimere il contesto.

Utilizzo

Attività

Ordina per

44

Classifica degli strumenti

Ranked by creation date, newest first; undated entries come last. I valori grezzi mantengono la propria finestra; una variazione richiede almeno due rilevazioni in 7 giorni.

  1. 1

    Skill
    Attivo

    The Agent Knowledge Kit & Architecture System — 10–32x token-efficient architecture discovery for AI coding agents

    TypeScript
    • agentic-ai
    • ai-agents
    • antigravity
    • bun
    • claude-code

    githubmomentum stimatoApri fonte ↗

    Installa git clone https://github.com/chama-x/quiv ~/.claude/skills/quiv

    3stelle GitHub—
  2. 2

    MCP
    Attivo

    Context optimization for LLM agents: tool registry, result masking, budgeting, compaction.

    RustLicenza: NOASSERTION

      mcpcrescita misurataApri fonte ↗

      Installa claude mcp add distil -- npx @munhq/distil

      1stelle GitHubstabile
    • 3

      Skill
      Attivo

      A Claude Code hook that routes each Read on a PDF to the path that actually works: text layer to UTF-8 text, scans to vision, and neither one through poppler.

      PythonLicenza: MIT
      • claude-code
      • claude-code-hooks
      • claude-code-plugin
      • pdf
      • poppler

      githubcrescita misurataApri fonte ↗

      Installa /plugin marketplace add WatsonTsai/pdf-text-router

      7stelle GitHubstabile
    • 4

      Skill
      Attivo

      Profile Claude Code context, discover repeated workflows, and regression-test your skills. Local-only, zero-dependency.

      PythonLicenza: MIT
      • agent-skills
      • anthropic
      • claude-code
      • claude-code-plugin
      • claude-code-plugins

      githubcrescita misurataApri fonte ↗

      Installa /plugin marketplace add nks0614/harnessay

      5stelle GitHub+2 (+66.7 %)
    • 5

      Skill
      Attivo

      Cut Claude Code token costs by delegating bulk LLM work to cheap models. 600+ models via NanoGPT, OpenRouter, Groq or local Ollama — batched, resumable, with real per-call cost tracking.

      PythonLicenza: MIT
      • ai-agents
      • batch-processing
      • cheap-llm
      • claude-code
      • claude-code-agents

      githubcrescita misurataApri fonte ↗

      Installa /plugin marketplace add bicced/sidequest

      1stelle GitHubstabile
    • 6

      MCP
      Attivo

      Lossless context compression: 2-8x fewer tokens, byte-exact recovery, search inside payloads

      PythonLicenza: MIT

        mcpcrescita misurataApri fonte ↗

        Installa claude mcp add densely -- uvx densely

        6stelle GitHubstabile
      • 7

        Skill
        Attivo

        Claude Code skill: caveman-style token efficiency, spoken with a real Spanish regional accent

        JavaScriptLicenza: MIT
        • agent-skills
        • ai-agents
        • anthropic
        • claude-code
        • claude-code-plugin

        githubcrescita misurataApri fonte ↗

        Installa /plugin marketplace add Mun1to/SpanishCaveMan

        2stelle GitHubstabile
      • 8

        CLI
        Attivo

        Measure and shrink the token cost of Claude Code skills, tier by tier: the always-on system-prompt tax, the per-invocation body, and on-demand references, each priced at the rate it is actually billed.

        PythonLicenza: MIT
        • claude
        • claude-code
        • claude-code-skills
        • cli
        • cost-optimization

        githubcrescita misurataApri fonte ↗

        1stelle GitHubstabile
      • 9

        MCP
        Attivo

        Statically audits MCP tool surfaces for token cost, schema quality, and design issues.

        TypeScript

          mcpcrescita misurataApri fonte ↗

          Server remoto da dichiarare nella configurazione MCP https://mcplint-web.vercel.app/api/mcp

          0stelle GitHubstabile
        • 10

          MCP
          Attivo

          Verifies token/cost savings claimed by AI context-reduction proxies via MCP tools.

          TypeScriptLicenza: Apache-2.0

            mcpcrescita misurataApri fonte ↗

            Installa claude mcp add tokentrust -- uvx tokentrust-cli

            0stelle GitHubstabile
          • 11

            Skill
            Attivo

            🪒 YAGNI enforcement at the harness level, nudges toward the simplest solution and gates against unnecessary dependencies, file sprawl, and unchecked build growth.

            JavaScriptLicenza: MIT
            • ai
            • anthropic
            • claude
            • claude-code
            • claude-code-plugin

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add V-Songbird/razor

            8stelle GitHub+1 (+14.3 %)
          • 12

            Skill
            Attivo

            🤫 Token-lean sessions at the harness level. An easy to grasp output style, output-shrinking hooks, and log compression cut both input and output tokens.

            JavaScriptLicenza: MIT
            • agent-skills
            • ai
            • anthropic
            • claude
            • claude-code

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add V-Songbird/hush

            47stelle GitHub+4 (+9.3 %)
          • 13

            Skill
            Attivo

            Lossless context compression plugin for Claude Code & OpenCode.

            TypeScriptLicenza: BSD-3-Clause
            • anthropic
            • chatgpt
            • claude
            • claude-code
            • claude-code-plugin

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add aerovato/magic-compact

            146stelle GitHub+1 (+0.69 %)
          • 14

            Agente
            Attivo

            Control plane for AI coding agents: route tasks, reduce token spend, run multi-agent workflows, fallback executors, and track cost per task.

            PythonLicenza: Apache-2.0
            • a2a
            • ag-ui
            • agentic-ai
            • ai-agents
            • ai-coding

            githubcrescita misurataApri fonte ↗

            16stelle GitHubstabile
          • 15

            Skill
            Attivo

            Eco mode for Claude Code. /eco: -31% to -73% output tokens with critical findings intact; /eco-max: up to -75% with lowered effort. Measured hardest on Claude Fable 5 (fable5), deep-studied on Sonnet 5, works on Opus 4.8 too. We publish our…

            JavaScriptLicenza: MIT
            • ai-agents
            • anthropic
            • claude
            • claude-ai
            • claude-code

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add sup3x/claude-code-eco

            33stelle GitHubstabile
          • 16

            Skill
            Attivo

            Staged context pruning for OpenCode, Oh My Pi, pi, and Claude Code.

            TypeScriptLicenza: AGPL-3.0
            • ai
            • ai-agents
            • claude-code
            • claude-code-plugin
            • coding-agent

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add AshishKumar4/better-compact

            12stelle GitHub+1 (+9.1 %)
          • 17

            Skill
            Attivo

            A production-ready Claude Code setup. Global CLAUDE.md, coding rules, per-project memory, and token optimization.

            ShellLicenza: MIT
            • ai-agents
            • claude
            • claude-code
            • claude-code-plugin
            • context-engineering

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add janmaaarc/basecamp

            7stelle GitHubstabile
          • 18

            Skill
            Attivo

            对标 claude-hud 核心功能，控制在 7 行以内更轻量。内置 13 个国产大模型平台用量查询 (智谱 GLM / MiniMax / DeepSeek / Kimi / StepFun / SiliconFlow 等)，第三方代理用户也能实时看到额度消耗。

            TypeScriptLicenza: MIT
            • anthropic
            • claude
            • claude-code
            • claude-code-plugin
            • cli

            githubcrescita misurataApri fonte ↗

            Installa /plugin marketplace add zander-zyx/claude-mini-hud

            9stelle GitHubstabile
          • 19

            MCP
            Attivo

            Cut context bloat in your AI-agent stack: find and safely prune unused skills, MCP servers and subagents from real transcript evidence

            GoLicenza: MIT
            • agent-skills
            • ai-agents
            • anthropic
            • claude-code
            • claude-code-plugin

            githubcrescita misurataApri fonte ↗

            55stelle GitHub+1 (+1.9 %)
          • 20

            MCP
            Attivo

            MCP server and proxy that compresses LLM prompts, tool output, and replies to cut token cost.

            ↓ 171 download/settimanaRustLicenza: MPL-2.0

              mcpcrescita misurataApri fonte ↗

              Installa claude mcp add llmtrim -- npx @llmtrim/cli

              222stelle GitHub+6 (+2.8 %)
            • 21

              Skill
              Attivo

              Claude Code skill — delegate coding tasks to Mistral Vibe, supervise via git diff & save 50-90% Claude token usage

              Shell
              • claude
              • claude-ai
              • claude-ai-skills
              • claude-code
              • claude-skill

              githubcrescita misurataApri fonte ↗

              Installa git clone https://github.com/pcx-wave/vibe-skill ~/.claude/skills/vibe-skill

              117stelle GitHubstabile
            • 22

              Skill
              Attivo

              Governance framework for AI coding agents. It runs them through a five-step workflow (plan, build, review, test, ship) where no step counts as done without evidence. Drop-in rules and guardrails for Claude Code, Codex, Cursor, Copilot, and…

              PythonLicenza: MIT
              • agent-framework
              • agent-orchestration
              • agentic-development
              • agents-md
              • ai-agent

              githubcrescita misurataApri fonte ↗

              Installa git clone https://github.com/KbWen/agentic-os ~/.claude/skills/agentic-os

              155stelle GitHub+8 (+5.4 %)
            • 23

              MCP
              Attivo

              A MCP that turns PDFs into structured text, tables, figures, metadata, annotations, and page-level evidence for selective retrieval by MCP-compatible AI agents. Improves the information quality AI agents receive while also reducing token…

              JavaScriptLicenza: AGPL-3.0
              • ai
              • ai-tools
              • claude
              • claude-code
              • claude-code-plugin

              githubcrescita misurataApri fonte ↗

              17stelle GitHubstabile

            Risorse didattiche e di riferimento

            Ranked by creation date, newest first; undated entries come last. Queste risorse restano accessibili separatamente e non partecipano alla classifica principale.

            1. 1

              mcp-agent-skills-hub

              MCP
              AttivoApri fonte ↗

              Molavi Agent Skills: Curated AI agent skills and MCP configs for Antigravity, Cursor, Codex & Claude — by Taghi Molavi

              JavaScriptLicenza: MIT
              • agentic-workflows
              • ai-agents
              • ai-skills
              • antigravity
              • awesome-list
              githubRisorsacrescita misurata
              8stelle GitHubstabile
            2. 2

              outsourcerer

              Skill
              AttivoApri fonte ↗

              Make the most out of your subscriptions. Delegate work to other harnesses and models, while keeping your main session the orchestrator. Nothing new to learn. Keep working like you do, but better.

              Shell
              • agent-skills
              • agentic-workflow
              • claude-code
              • codex
              • fable
              github

              Installa git clone https://github.com/alexgreensh/outsourcerer ~/.claude/skills/outsourcerer

              Risorsacrescita misurata
              142stelle GitHub+3 (+2.2 %)