- Home
- Skills
- Prompt & Skill Engineering
- Terse Mode: Shorter Answers, Fewer Tokens
Works with the AI Agents you already use
Terse Mode: Shorter Answers, Fewer Tokens
Here's what I'll do...", recaps of what the tool output already shows, "Let me know if you'd like..." - verbose agents waste tokens, money and your reading time.
Free
Terse Mode: Shorter Answers, Fewer Tokens
Example session with this skill installed
terse short. Why does my Docker build take 6 minutes every time even when I only change one line of code?
- Read your context and instructions
- Compiled the terse mode
Cause: COPY . . before RUN npm ci → any code change invalidates the deps layer → full reinstall.
Fix — copy manifests first, code last:
FROM node:22-alpine
WORKDIR /app
COPY package.json package-lock.json ./
RUN npm ci
COPY . .
RUN npm run build
Also
- Add
.dockerignore:node_modules,.git,dist,.env - BuildKit cache for npm:
RUN --mount=type=cache,target=/root/.npm npm ci
Expected: code-only change → deps layer cached, build ≈ build step only.
Terse short: 75 words, no preamble or recap. Dockerfile and commands unchanged.
Connects securely to your tools. The creator never sees your data.
About this skill
Less talk, same answer
"Great question! Here's what I'll do...", recaps of what the tool output already shows, "Let me know if you'd like..." - verbose agents waste tokens, money and your reading time. Terse Mode switches your agent to a compact style that keeps every fact.
What it does
- Three levels you can switch anytime ("terse lite", "terse short", "terse ultra", "normal mode" to exit):
- lite: full sentences, no filler or recaps (~30-40% shorter)
- short (default): fragments, bullets, arrows and common abbreviations (~50-60%)
- ultra: telegraphic, result + next action only (~65-75%)
- Always cuts: openers, restating the request, narrating tool use, recaps, hedging stacks, unrequested alternatives and theory, sign-off questions.
- Never cuts: code, commands, paths, numbers, versions, error messages and names (kept verbatim), real risk warnings (data loss, security, money, legal), questions it needs answered, ordered steps.
- Smart exceptions: switches to full detail when you ask "why" or "explain", when you seem new to a topic, for destructive instructions, and for deliverables written for other people (docs, emails, PRs, reports) - terse is for chat, not artifacts.
- Worked example with word counts: 74 -> 20 -> 16 -> 8 words for the same correct answer.
- Measure: compares normal vs terse word counts on recent replies when you ask.
Pairs with Cut Agent Token Costs (caching, routing, budget guards) for the full token-saving stack.
Includes
- SKILL.md - levels, cut/keep rules, exceptions, examples
Works with Claude Code, Claude (Cowork), Codex, Cursor, Copilot, Gemini CLI and any SKILL.md-compatible agent.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.
Terse style only applies to chat interactions. It will not compress documents, PR descriptions, or user-facing reports meant for third parties.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
This skill has been used 10 times.
Trust & safety
Security scan
Security scan passed 8 days ago
- Free with an Agensi account