Token Efficiency

The only honest metric for an efficiency skill is the bill. Every verdict here comes with a measured before/after: token counts and cache-hit rates from real workloads, not vibes about "optimized prompts."

80 skills listed · 36 fully verified · top 10 ranked →

  1. 001

    Kill Dev Process

    Classifies and safely kills orphaned dev servers/browsers with a hard exclusion list for databases and the IDE.

    by nokonoko1203 tested Jul 13, 2026

    9.6/10 Tested · Works
  2. 002

    Analyze Costs

    Cloud cost analysis and savings via Harness CCM, plus anomaly and commitment checks.

    by harness tested Jul 14, 2026

    9.2/10 Tested · Works
  3. 003

    Checkpoint

    Writes a structured, resumable session-handoff snapshot with file:line pointers and a resume prompt.

    by pedrohcgs tested Jul 20, 2026

    9.2/10 Tested · Works
  4. 004

    Free My CPU

    Read-only CPU/load/Docker audit workflow with safety-tiered remediation, never kills anything unasked.

    by wquguru tested Jul 13, 2026

    9.2/10 Tested · Works
  5. 005

    Link Ticket To Session

    Link a Claude Code session to a Linear/Jira/GitHub ticket or PR in the karma dashboard

    by JayantDevkar tested Jul 31, 2026

    9.2/10 Tested · Works
  6. 006

    Mcp Client

    CLI to list and call tools on any MCP server without loading all schemas into context

    by coleam00 tested Jul 21, 2026

    9.2/10 Tested · Works
  7. 007

    Skill Tree Generator

    Converts a bloated Claude skill into a routing tree that loads only what each request needs.

    by maipianworni tested Jul 14, 2026

    9.2/10 Works with setup
  8. 008

    Codex Autoresearch

    Measured, resumable optimization loop with baseline, correctness gate, and evidence ledger

    by TheGreenCedar tested Jul 21, 2026

    8.8/10 Tested · Works
  9. 009

    Compact Guard

    Pre/post-compaction checklist that preserves task state inside Claude Code's real compaction limits.

    by rohitg00 tested Jul 14, 2026

    8.8/10 Tested · Works
  10. 010

    Context Optimizer

    Concrete thresholds and commands for managing Claude's context window and token budget

    by rohitg00 tested Jul 20, 2026

    8.4/10 Tested · Works
  11. 011

    Independent Research

    Stops guessed fixes: reproduce the error and verify with bash before recommending

    by NTCoding tested Jul 31, 2026

    8.4/10 Tested · Works
  12. 012

    Agent Research Aggregator

    Scans AI agent cache logs and formats them into PaperOrchestra idea.md + experimental_log.md

    by Ar9av tested Jul 21, 2026

    8.0/10 Tested · Works
  13. 013

    Cost Tracker

    Static reference for what drives Claude Code session cost and per-task-type budget thresholds — not a transcript auditor.

    by rohitg00 tested Jul 14, 2026

    8.0/10 Tested · Works
  14. 014

    Prompt Discipline

    Diagnose the failure first, then make the smallest fix — no magic-template rewrites.

    by SkillProof tested Jul 11, 2026

    8.0/10 Tested · Works
  15. 015

    Token Discipline

    Input-side token rules for Claude Code — ~20% off multi-step work, wash on one-shots.

    by SkillProof tested Jul 11, 2026

    8.0/10 Tested · Works
  16. 016

    Agentdb Memory Patterns

    Persistent agent memory patterns and CLI workflow built on the AgentDB vector store

    by spencermarx tested Jul 31, 2026

    7.6/10 Works with setup
  17. 017

    Hivemind Graph

    AST code graph you query as files — callers, imports, impact, no build

    by activeloopai tested Jul 20, 2026

    7.6/10 Works with setup
  18. 018

    Prompt Caching

    Cross-provider (Anthropic/OpenAI/Bedrock/Gemini) prompt-caching diagnostic playbook — four of its own linked reference files are missing from the repo.

    by cobusgreyling tested Jul 14, 2026

    7.2/10 Works with setup
  19. 019

    Golang Benchmark

    Statistically rigorous Go benchmarking, pprof profiling, and benchstat regression workflow

    by samber tested Jul 10, 2026

    9.2/10 Tested · Works
  20. 020

    SkillOpt-Sleep

    Offline nightly sleep-cycle that mines past sessions and gates CLAUDE.md/SKILL.md edits.

    by microsoft tested Jul 10, 2026

    9.2/10 Tested · Works
  21. 021

    Token Doctor

    Reads your local Claude Code/Cowork transcripts and prints a real cost-shape diagnosis in the terminal.

    by techwolf-ai tested Jul 13, 2026

    9.2/10 Tested · Works
  22. 022

    Claude Usage

    Ground-truth Claude token/cost reports by parsing JSONL directly, catching what ccusage misses.

    by tdimino tested Jul 14, 2026

    8.8/10 Tested · Works
  23. 023

    Delegation Triggers

    Explicit rules for when to spawn subagents vs. work directly, tuned against a 408-run benchmark.

    by vitaliikapliuk tested Jul 14, 2026

    8.8/10 Tested · Works
  24. 024

    Compress Prompt

    Strips a verbose prompt down to the minimal instruction a model actually needs, with a keep/drop rubric.

    by doodledood tested Jul 14, 2026

    8.4/10 Works with setup
  25. 025

    Memory Recall

    Forked subagent that recalls past-session decisions via the memsearch semantic-memory CLI.

    by zilliztech tested Jul 20, 2026

    7.6/10 Works with setup
  26. 026

    memU

    Gives the agent persistent cross-session memory to save and recall project context.

    by NevaMind-AI tested Jul 12, 2026

    7.6/10 Works with setup
  27. 027

    Aws Cost Operations

    AWS cost analysis and CloudWatch alarm patterns; MCP servers need plugin install + creds

    by zxkane tested Jul 31, 2026

    7.2/10 Works with setup
  28. 028

    Discover

    Instruments a repo for evo's autonomous optimization: benchmark, baseline, first run

    by evo-hq tested Jul 20, 2026

    7.2/10 Works with setup
  29. 029

    Paper Finder

    Multi-angle paper search into a persistent memory bank and BibTeX

    by bchao1 tested Jul 31, 2026

    7.2/10 Works with setup
  30. 030

    Simulink Single Precision Conversion

    Convert double-precision Simulink models to single precision via Fixed-Point Designer

    by matlab tested Jul 21, 2026

    6.4/10 Works with setup
  31. 031

    Add Backend

    Scaffold and wire a backend template (auth, ORM, CRUD) into the dashboard frontend.

    by ahpxex

    In test queue
  32. 032

    Code Execution

    Run local Python for bulk code ops, returning summaries instead of full source

    by mhattingpete tested Jul 21, 2026

    Tested · Didn't pass
  33. 033

    Hivemind Goals

    Creates and tracks team goals, KPIs, and tasks through the Hivemind virtual filesystem.

    by activeloopai tested Jul 10, 2026

    Tested · Didn't pass
  34. 034

    Market Session Tracker

    Monitor a stock watchlist live across pre-market, open, intraday, and close sessions.

    by Innei

    In test queue
  35. 035

    Matrix Deep Research

    Conduct multi-source research across web, docs, and memory for a comprehensive report.

    by ojowwalker77

    In test queue
  36. 036

    Obsidian

    Guides Obsidian.md plugin development with ESLint rules and API usage.

    by gapmiss

    In test queue
  37. 037

    Performance Profiling

    Profiles CPU, memory, network, and energy usage in iOS apps with Instruments.

    by conorluddy

    In test queue
  38. 038

    Pr Precheck

    Checks a repo's PR history for prior attempts before opening a new pull request.

    by vouchdev

    In test queue
  39. 039

    Wp Performance Review

    Reviews WordPress PHP code for performance issues and scalability anti-patterns.

    by elvismdev

    In test queue
  40. 040

    WJS Burning Subtitles

    Burns SRT subtitles into video for WeChat/Douyin or soft-muxes a togglable track, with libass auto-fallback.

    by jianshuo tested Jul 13, 2026

    8.8/10 Tested · Works
  41. 041

    ACM Config

    Interactive Q&A that writes a validated ~/.claude/claudikins-acm.conf for the Claudikins context-handoff tool.

    by elb-pr tested Jul 16, 2026

    9.6/10 Tested · Works
  42. 042

    Agent Prompt Engineering

    A five-part checklist (role, paths, deliverables, constraints, comms) for writing sub-agent prompts that need zero follow-up.

    by hesreallyhim tested Jul 15, 2026

    9.6/10 Tested · Works
  43. 043

    Clean Cache

    Tiered, confirmation-gated cache cleanup for Flutter/Android/iOS/Node workspaces — verified not to touch lockfiles or source.

    by addxai tested Jul 14, 2026

    9.6/10 Tested · Works
  44. 044

    Perfetto Performance Analysis

    Perfetto trace analysis with a pinned trace_processor and 235 query graphs

    by Gracker tested Aug 10, 2026

    9.6/10 Tested · Works
  45. 045

    Tidy Skill

    Read-only hygiene audits that stop agents littering repos with plan.md/todo.md junk.

    by Phoenix0531-sudo tested Jul 16, 2026

    9.6/10 Tested · Works
  46. 046

    Agentic Coach

    Scores a vague prompt on 6 dimensions and asks Socratic questions instead of silently rewriting and running it.

    by cgbarlow tested Jul 15, 2026

    9.2/10 Tested · Works
  47. 047

    Eco Max

    Frugality rules plus a low reasoning-effort override for routine chores — measured, not asserted.

    by sup3x tested Jul 16, 2026

    9.2/10 Tested · Works
  48. 048

    Optimise Skill

    Vendors Microsoft's SkillOpt/OPRO engine to retrain one skill's judge-shaped rubric against a labelled outcome corpus, gated by a held-out set.

    by yungbose tested Jul 15, 2026

    9.2/10 Tested · Works
  49. 049

    ADB Android Control

    ADB command cheatsheet plus a real, 349-test Python CLI for Android device automation from Claude.

    by hah23255 tested Jul 16, 2026

    8.8/10 Tested · Works
  50. 050

    Brain Uninstall

    Confirmation-gated, password-masking teardown for the mybrain memory plugin across Docker, local, and remote Postgres.

    by robertsfeir tested Jul 16, 2026

    8.8/10 Works with setup
  51. 051

    Fable Commander

    Maker/verifier loop with hard gates, 3-round caps and human stops before commits

    by DennisWei9898 tested Aug 5, 2026

    8.8/10 Tested · Works
  52. 052

    Fable Fanout

    Forces parallel one-message delegation instead of serial spawn-wait-spawn on multi-unit tasks.

    by rennf93 tested Jul 15, 2026

    8.8/10 Tested · Works
  53. 053

    Fable God Mode

    10-80-10 delegation discipline: expensive model plans and reviews, cheap subagents do the verbose middle.

    by nagarjuna-msr tested Jul 16, 2026

    8.8/10 Tested · Works
  54. 054

    Memory Loop

    Structured project-memory writer for .claude/memory/MEMORY.md — but ships as a plugin, not a bare skill.

    by yucai0302 tested Jul 14, 2026

    8.8/10 Works with setup
  55. 055

    Prompt Enhancer

    Turns a vague '-e'/'--enhance'-flagged request into a structured INTENT/ACTION spec using session memory.

    by Prorise-cool tested Jul 15, 2026

    8.8/10 Works with setup
  56. 056

    Prompt Sensei

    Scores your prompts to coding agents by stage and coaches one habit at a time

    by chengzhongwei tested Jul 17, 2026

    8.8/10 Tested · Works
  57. 057

    Vaultr Memory

    Extracts identity/preferences/goals/beliefs/people/state from Vaultr notes with decay-based memory files.

    by skoowoo tested Jul 15, 2026

    8.8/10 Works with setup
  58. 058

    WSL Interop

    Windows/WSL path conversion, clickable file links, and SSH-agent interop rules for Claude Code in WSL.

    by JeremiahChurch tested Jul 15, 2026

    8.8/10 Tested · Works
  59. 059

    Gridhand

    Rust CLI that lets an agent click desktop GUIs by naming a labeled grid cell instead of guessing pixel coordinates.

    by ZachRouan tested Jul 16, 2026

    8.4/10 Works with setup
  60. 060

    Mythos Scaffold

    Agentic-discipline scaffold: decomposition, verification, stuck-detection, plus research and migration domain modes.

    by kasparovabi tested Jul 16, 2026

    8.4/10 Tested · Works
  61. 061

    Agent Team Design

    Design multi-agent teams with explicit memory, policy, and eval owners

    by agentlas-ai tested Jul 17, 2026

    8.0/10 Tested · Works
  62. 062

    Parable

    Working-style rules that move agent deliberation out of chat into a notes file

    by SamBWarren tested Jul 29, 2026

    8.0/10 Works with setup
  63. 063

    Browser CDP

    Drive Chrome over CDP to reuse existing login sessions for automation

    by worldwonderer tested Jul 17, 2026

    7.6/10 Works with setup
  64. 064

    Continue (Token Saver)

    Zero-token session context restore by reading transcripts directly, not /compact

    by ww-w-ai tested Jul 17, 2026

    7.6/10 Works with setup
  65. 065

    Fable Advisor

    Escalate hard calls to a Fable subagent, with a live main-model probe

    by echo-of-machines tested Aug 5, 2026

    7.6/10 Works with setup
  66. 066

    FewWord

    Auto-offloads big command output to files, returns a compact pointer

    by sheeki03 tested Jul 17, 2026

    7.6/10 Works with setup
  67. 067

    Handoff DS

    Offloads execution-heavy coding tasks to DeepSeek in the background to save quota

    by dazuiba tested Jul 20, 2026

    7.6/10 Works with setup
  68. 068

    MeMesh Review

    Health-scores and cleans up your MeMesh AI memory database

    by PCIRCLE-AI tested Jul 16, 2026

    7.6/10 Works with setup
  69. 069

    Recursive Decomposition

    Large-context tactics: grep-first, chunk, recurse via sub-agents

    by massimodeluisa tested Jul 17, 2026

    7.6/10 Works with setup
  70. 070

    Sandy

    Deterministic replay of MCP/tool-call sequences so repeated workflows skip LLM inference entirely.

    by Sangkwun tested Jul 16, 2026

    7.6/10 Works with setup
  71. 071

    STRETCH AI Yoga

    Self-applied cognitive practices an AI agent runs on itself between tasks

    by stretchvancouver tested Jul 18, 2026

    7.6/10 Works with setup
  72. 072

    Synapptic

    Mines your Claude Code session transcripts into a living user/AI-failure profile and guard list

    by appcuarium tested Jul 15, 2026

    7.6/10 Works with setup
  73. 073

    Claude Mind

    Persistent cross-session memory for Claude stored in one portable .mv2 file

    by memvid tested Jul 18, 2026

    7.2/10 Works with setup
  74. 074

    Compute Env Setup

    Playbook for standing up remote GPU/conda/Slurm/Modal compute environments

    by xuzhougeng tested Jul 18, 2026

    7.2/10 Works with setup
  75. 075

    Token Reducer

    RAG context-compression pipeline for agents — its CLI script isn't in the skill folder.

    by Madhan230205 tested Jul 14, 2026

    6.8/10 Works with setup
  76. 076

    Oracle Challenge

    Structured critical-thinking framework to pressure-test a decision before you commit.

    by martinffx tested Jul 15, 2026

    6.4/10 Works with setup
  77. 077

    AVIZ Skills Installer

    Installs skills from the AVIZ library — but its own doc/discovery links point to a dead domain.

    by aviz85 tested Jul 14, 2026

    Tested · Didn't pass
  78. 078

    Coderlm

    Turns codex/gemini/claude into a recursive agent with bundled bash output truncation guard

    by CyrusNuevoDia tested Jul 20, 2026

    Tested · Didn't pass
  79. 079

    Outsourcerer

    Offload grunt work to a cheaper engine while Claude orchestrates

    by alexgreensh tested Jul 17, 2026

    Tested · Didn't pass
  80. 080

    Pull Workflow (Hive)

    Pulls a crowd-sourced multi-step browser procedure from Hive and executes it step by step.

    by Brainrot-Creations tested Jul 15, 2026

    Tested · Didn't pass

Can't find what you need?

Request a skill — we'll test or build it

Tell us the job. Within a few days you get back a link to a tested skill that already does it, a SKILL.md you can build from, or a ready-to-use prompt. Popular asks get built for the catalog.