skills.rest
IsKenKenYa
IsKenKenYaCOMMUNITY
Explore all skills by this author →
GitHub

writing-for-agents

Write and structure documents that AI agents consume, including skills and AGENTS.md files.

2|Updated Jun 21, 2026
Search Tags:#documentation#prompt engineering#ai agents#skill authoring#progressive disclosure#agents.md#context pointers
One-click install
npx skills add https://github.com/IsKenKenYa/skills --skill writing-for-agents-iskenkenya
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-for-agents
Source: https://github.com/IsKenKenYa/skills/tree/main/skills/productivity/writing-for-agents
Command: npx skills add https://github.com/IsKenKenYa/skills --skill writing-for-agents-iskenkenya

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Documents written for AI agents often fail unpredictably: agents miss referenced material, rush steps, or get lost in bloated files. This Skill provides a systematic reference for writing agent-facing documents—skills, AGENTS.md, CLAUDE.md—so the agent follows the same process reliably on every run. ## Core Features & Use Cases - Context pointer design: Write descriptions and pointer lines whose wording controls when the agent reaches referenced material, with trigger branches front-loaded. - Information hierarchy management: Apply progressive disclosure, co-location, and pruning rules to decide what stays inline versus behind a pointer, avoiding document sprawl and sediment. - Skill mechanics guidance: Choose between model-invoked and user-invoked skills, design router skills, and split documents by sequence or invocation (via SKILL-MECHANICS.md). - Use Case: When creating a new skill or editing an AGENTS.md file, use this Skill to sharpen completion criteria, replace vague prose with leading words, and eliminate no-op instructions that add context load without changing behavior. ## Quick Start Use the writing-for-agents skill to review and improve the skill document I am drafting.

Frequently Asked Questions about writing-for-agents

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

FAQPage Schema
How do I write a good skill description for an AI agent?▼

Write the description as a context pointer: front-load the leading trigger word and list one trigger per distinct branch the skill handles. Cut synonyms and identity the body already carries, since every word of an always-loaded description costs context on every turn.

What is the difference between model-invoked and user-invoked skills?▼

A model-invoked skill keeps a description so the agent can fire it autonomously, paying permanent context load. A user-invoked skill sets disable-model-invocation: true, removing it from the agent's reach so only a human typing its name can trigger it, with zero context load.

When should I split a skill into multiple documents?▼

Split by sequence when visible later steps tempt the agent to rush the current one, and split by invocation when a distinct trigger word deserves its own model-invoked skill. Each split spends context or cognitive load, so the cut must earn it.

How do I keep an AGENTS.md file from getting too long?▼

Apply progressive disclosure: push reference material only some branches need into separate files behind context pointers, keeping inline what every branch needs. Prune stale lines, duplicated meanings, and no-op instructions the model already follows by default.

Why should I avoid negative instructions in agent prompts?▼

Negation drags the forbidden behavior into context and makes it more available, since the strongly-activated concept overruns the weak negation. State the positive target behavior instead, reserving prohibitions for hard guardrails paired with a positive alternative.

Related Skills

Explore More Software Engineering Skills→
xiechyxiechy
3

context-initialization

Create or update AGENTS.md to enforce search-first usage of existing skills.

Community
Basic
openchamberopenchamber
9.5k

writing-for-agents

Guides writing and editing of agent-facing documents such as skills and AGENTS.md files.

Official
Intermediate
DarkCodePEDarkCodePE

skill-stocktake

Audit Claude skills and commands against a quality checklist.

Community
Advanced
s4kr4s4kr4
1

py-formatter

Standardize Python formatting and linting with Black, flake8, isort, and mypy.

Community
Intermediate