token-efficiency · nickcrew — Claude Code skill: compressed agent language
token-efficiency — a Claude Code skill that switches the agent to compact notation — symbols and abbreviations instead of full sentences. Kicks in when the context window is nearly full or a short status line is enough.
MIT 245 installs ★ 71 updated 2026-06-29
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a claude-codeWhat token-efficiency does
Installs into Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. The skill defines a symbol dictionary in three groups: logic and flow (→, ⇒, ∴, ∵), statuses (✅, ❌, ⚠️, 🔄, ⏳), and domains (⚡ performance, 🔍 analysis, 🛡️ security, 🏗️ architecture).
Frequent terms get short forms: cfg, impl, deps, val, perf. A sentence like “the authentication system has a security vulnerability in the user validation function” becomes auth.js:45 → 🛡️ sec risk in user val().
Inside there is only a SKILL.md with symbol tables and examples. The agent switches to compressed mode when context is above 75% full, during large codebase analysis, in multi-step workflows, or when the user asks for brevity. Emoji statuses render inconsistently across terminals and fonts — some will show as boxes.
What it helps with
- shrink agent replies when context is almost full
- report build, test and deploy status in one line
- tag a task type with a domain symbol
- replace frequent terms with short abbreviations
- keep one compressed vocabulary across a team
Example prompts for the agent
- Compress replies using the token-efficiency symbol system
- Give me the build status as build ✅ ” test 🔄 ” deploy ⏳
- Context is at 80%, switch to token-efficiency notation
How to install token-efficiency
1Skills CLI (Vercel)
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a claude-code2Manually: put the files in place
~/.claude/skills/token-efficiency/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/token-efficiency/
1Skills CLI (Vercel)
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a codex2Manually: put the files in place
.agents/skills/token-efficiency/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a cursor2Manually: put the files in place
.cursor/skills/token-efficiency/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install NickCrew/Claude-Cortex skills/token-efficiency2Manually: put the files in place
.github/skills/token-efficiency/SKILL.md1Skills CLI (Vercel)
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a gemini-cli2Platform CLI
gemini skills install https://github.com/NickCrew/Claude-Cortex --path skills/token-efficiencyManually: put the files in place
.devin/skills/token-efficiency/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about token-efficiency
How do I install token-efficiency in Claude Code?
Run npx skills add NickCrew/Claude-Cortex --skill token-efficiency -a claude-code. The skill file lands in ~/.claude/skills/token-efficiency/SKILL.md.
What does token-efficiency do?
It introduces a compressed notation: symbols for logic and status, plus abbreviations like cfg and impl. The agent falls back to it when context is tight or short answers are needed.
When does the compressed mode actually activate?
Per the skill conditions: context above 75% full, large codebase analysis, complex multi-step work, or an explicit request from the user for brevity.
Source text
SKILL.md · MIT
--- name: token-efficiency description: Compressed communication using symbols and abbreviations. Use when context is limited or brevity is needed. keywords: - compressed communication - abbreviations - token - efficiency - token efficiency --- # Token Efficiency Compressed communication for limited context windows. ## Symbol System ### Logic & Flow | Symbol | Meaning | |--------|---------| | → | leads to, implies | | ⇒ | transforms to | | ← | rollback | | & | and | | \| | or | | » | sequence/then | | ∴ | therefore | | ∵ | because | ### Status | Symbol | Meaning | |--------|---------| | ✅ | complete/pass | | ❌ | failed/error | | ⚠️ | warning | | 🔄 | in progress | | ⏳ | pending | ### Domains | Symbol | Domain | |--------|--------| | ⚡ | performance | | 🔍 | analysis | | 🛡️ | security | | 🏗️ | architecture | ## Abbreviations - `cfg` config - `impl` implementation - `deps` dependencies - `val` validation - `perf` performance - `sec` security - `err` error ## Examples **Standard:** > "The authentication system has a security vulnerability in the user validation function" **Compressed:** > `auth.js:45 → 🛡️ sec risk in user val()` **Standard:** > "Build completed, now running tests, then deploying" **Compressed:** > `build ✅ » test 🔄 » deploy ⏳` ## When to Use - Context >75% full - Large codebase analysis - Complex multi-step workflows - User requests brevity
© NickCrew · MIT · NickCrew/Claude-Cortex
Similar skills
All skills for Agent setup: memory, context, orchestration →token-efficiency by other authors comes first.