skills.rest
klh
klhCOMMUNITY
Explore all skills by this author →
Website𝕏@klhougesenGitHub

all-skills

Guides agent output verbosity, efficiency, and formatting rules for any user request.

1|Updated Apr 16, 2026
Search Tags:#agent behavior#token efficiency#hallucination prevention#output formatting#verbosity#prompt rules#coding agents
One-click install
npx skills add https://github.com/klh/skills --skill all-skills-klh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: all-skills
Source: https://github.com/klh/skills/tree/main/.well-known/agent-skills/all-skills
Command: npx skills add https://github.com/klh/skills --skill all-skills-klh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents often produce verbose, unstructured, or inconsistent output that wastes tokens and breaks downstream parsing. This Skill enforces global behavior rules covering verbosity, efficiency, output format, and hallucination prevention across every agent response. ## Core Features & Use Cases - Output Discipline: Enforces structured output (JSON, bullets, tables), ASCII-only characters, and no conversational filler or status narration. - Efficiency Rules: Limits redundant file reads, caps tool calls at 50, and requires reading before writing to avoid wasted iterations. - Hallucination Prevention: Forbids inventing file paths, API endpoints, or field names; unknown values must return null or "UNKNOWN". - Use Case: Apply as a global behavior modifier when running coding agents, multi-agent systems, bots, or scheduled tasks where parseable, token-efficient output matters. ## Quick Start Ask the agent to apply the all-skills behavior rules so every response is concise, structured, and free of invented values.

Frequently Asked Questions about all-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make AI agent output more concise and structured?▼

Apply behavior rules that require structured output formats like JSON, bullets, or tables and prohibit prose unless a human reads it. This Skill enforces those rules globally so every response is parseable without post-processing.

How to prevent AI agents from hallucinating file paths and APIs?▼

Instruct the agent to never invent file paths, endpoints, function names, or field names, and to return null or "UNKNOWN" when a value is not verified. This Skill encodes exactly those hallucination prevention rules.

Can this skill work with Claude Code, Cursor, and other coding agents?▼

Yes, it follows the agentskills.io specification and installs via npx skills add, making it compatible with Claude Code, Codex, Cursor, Windsurf, Gemini CLI, and other agents supporting the standard.

What are the limitations of a global behavior modifier skill?▼

It only guides style and discipline; it does not perform domain tasks like debugging or testing itself. User instructions always override its rules, and it cannot enforce behavior outside the agent's context window.

Why does the skill restrict output to ASCII characters only?▼

ASCII-only output avoids smart quotes, em dashes, and ellipsis characters that can break JSON serialization or downstream parsers. This keeps agent responses safe for automated pipelines without escaping issues.

Related Skills

Explore More Software Engineering Skills→
nthpaulnthpaul

create-rule

Create Cursor rules with YAML frontmatter stored as .mdc files.

Community
Intermediate
Liuchun-ossLiuchun-oss
15

caveman

Reduce verbose prompts and responses by enforcing caveman-style brevity.

Community
Intermediate
issdandavisissdandavis
6

skill-synthesis

Combine multiple installed skills into an ordered execution stack.

Community
Advanced
misty-stepmisty-step
2

harness

Coordinate and enforce cross-harness discipline for agent skills and infrastructure.

Official
Advanced