Token efficiency: tools for developers

Tools that reduce token costs, compress context or cache LLM requests. An entry can appear under several use cases when its topics or description provide several explicit signals.

Use case

Activity

Sort by

96 entries in this view.

Tool ranking

Mixed ranking: measured growth takes priority; entries without two snapshots are estimated. Raw values keep their source-specific window, shown under each tool; a change requires at least two snapshots from the last 7 days.

  1. 1
    Active

    Reduces AI agent token usage by 90% via context compression and task checkpoint persistence.

    mcpmeasured growthOpen source ↗

    Install claude mcp add smart-context-mcp -- npx smart-context-mcp

    4GitHub starsstable
  2. 2
    Active

    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…

    githubmeasured growthOpen source ↗

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

    33GitHub starsstable
  3. 3
    Active

    Compact, source-linked context packs that reduce token waste for coding agents.

    mcpmeasured growthOpen source ↗

    Install claude mcp add capsule -- npx capsulectx

    4GitHub starsstable
  4. 4
    Active

    Rolling context compression for Claude Code — never hit the context wall. Auto-compresses old messages while keeping recent context verbatim. Zero config, zero latency. Works as a Claude Code plugin.

    githubmeasured growthOpen source ↗

    Install /plugin marketplace add NodeNestor/claude-rolling-context

    31GitHub starsstable
  5. 5
    Active

    20 power-user skills for Claude Code: session memory and consolidation, context compression, multi-agent orchestration, adversarial bug hunting, security review, calibrated estimation and decision archaeology. Drop-in SKILL.md files. Built…

    githubmeasured growthOpen source ↗

    Install git clone https://github.com/irfad7/claude-power-skills ~/.claude/skills/claude-power-skills

    4GitHub starsstable
  6. 6
    Active

    Restore prior Claude Code AND Codex sessions with zero LLM calls. Cross-tool session handoff, cache-expiry prevention, real cost dashboard. One plugin, both hosts.

    githubmeasured growthOpen source ↗

    Install /plugin marketplace add ww-w-ai/super-token-saver

    31GitHub starsstable
  7. 7
    Active

    Automatically detect and disable irrelevant Claude Code skills per project to save tokens and streamline your tech workflow.

    githubmeasured growthOpen source ↗

    Install git clone https://github.com/mrdenox109-nyx/skill-manager ~/.claude/skills/skill-manager

    5GitHub starsstable
  8. 8
    Dormant

    Give your Claude Code Agent Teams a memory. Auto-injects role-specific context into every new teammate — your team never starts blind again.

    githubmeasured growthOpen source ↗

    Install /plugin marketplace add Gr122lyBr/claude-teams-brain

    26GitHub starsstable
  9. 9
    Active

    Resume Claude Code work after rate/usage/context limits without replaying the prior transcript. Auto-saves at 90%/95% usage. Plugin-installable, 10 languages.

    githubmeasured growthOpen source ↗

    Install /plugin marketplace add sofumel/claude-handoff-revive

    6GitHub starsstable
  10. 10
    Active

    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.

    githubmeasured growthOpen source ↗

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

    7GitHub starsstable
  11. 11

    Skill
    Active

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

    githubmeasured growthOpen source ↗

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

    3GitHub starsstable
  12. 12
    UnknownLast commit: unavailable

    AI agent token-cost telemetry + 429 prediction. Per-agent attribution, anomaly + routing + quota.

    mcpmeasured growthOpen source ↗

    Install claude mcp add openclaw-cost-tracker-mcp -- uvx openclaw-cost-tracker-mcp

    129downloads over 7 daysstable
  13. 13
    UnknownLast commit: unavailable

    MCP server for Skim code transformation. Compresses code 60-95% for LLM context optimization.

    mcpestimated momentumOpen source ↗

    Install claude mcp add skim-mcp-server -- npx skim-mcp-server

    21downloads over 7 days
  14. 14

    Skill
    Active

    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.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add bicced/sidequest

    1GitHub stars
  15. 15
    Dormant

    Claude Code plugin — inventory all your installed skills, check structure, and find updates. Free, zero dependencies.

    githubestimated momentumOpen source ↗

    Install /plugin marketplace add VersoXBT/skill-manager

    6GitHub stars

Learning resources

Ranked by measured growth, normalized across sources. These resources remain available separately and do not take part in the main tool ranking.

  1. 1
    ActiveOpen source ↗

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

    githubResourcemeasured growth
    8GitHub starsstable
  2. 2
    ActiveOpen source ↗

    A cheat sheet that stops AI agents from scanning your whole repo and saves you tokens.

    github

    Install git clone https://github.com/toshon-jennings/SUMMARY-md ~/.claude/skills/SUMMARY-md

    Resourcemeasured growth
    0GitHub starsstable

Listed without a public metric

Mixed ranking: measured growth takes priority; entries without two snapshots are estimated. Raw values keep their source-specific window, shown under each tool; a change requires at least two snapshots from the last 7 days.

  1. 1
    UnknownLast commit: unavailable

    Semantic context compression and cognitive memory layer for LLM swarms.

    mcpmetric not publishedOpen source ↗

    Remote server, to be declared in the MCP configuration https://recallmax-mcp.vercel.app/api/mcp

    No public metric provided
  2. 2

    MCP
    UnknownLast commit: unavailable

    Pre-injection context compression for coding agents. Zero LLM calls, zero telemetry, offline-safe.

    mcpmetric not publishedOpen source ↗

    No public metric provided
  3. 3
    UnknownLast commit: unavailable

    LLM caching proxy (x402 USDC on Base) - exact + semantic cache. Free health.

    mcpmetric not publishedOpen source ↗

    Remote server, to be declared in the MCP configuration https://cache.api.ainode.tech/mcp

    No public metric provided
  4. 4
    UnknownLast commit: unavailable

    Cut LLM token costs: count tokens, estimate cost, slim prompts, and pick the cheapest capable model.

    mcpmetric not publishedOpen source ↗

    Install claude mcp add mcp-token-optimizer -- npx mcp-token-optimizer

    No public metric provided