kramme:agents-md
This skill should be used when the user asks to "update AGENTS.md", "add to AGENTS.md", "maintain agent docs", or needs to add guidelines to agent instructions. Guides discovery of local skills and enforces structured, keyword-based documentation style.
Best use case
kramme:agents-md is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
This skill should be used when the user asks to "update AGENTS.md", "add to AGENTS.md", "maintain agent docs", or needs to add guidelines to agent instructions. Guides discovery of local skills and enforces structured, keyword-based documentation style.
Teams using kramme:agents-md should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/kramme-agents-md/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How kramme:agents-md Compares
| Feature / Agent | kramme:agents-md | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
This skill should be used when the user asks to "update AGENTS.md", "add to AGENTS.md", "maintain agent docs", or needs to add guidelines to agent instructions. Guides discovery of local skills and enforces structured, keyword-based documentation style.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Adding to AGENTS.md
AGENTS.md is the canonical agent-facing documentation. Each rule should be minimal, but many rules are OK.
## Before Writing
Discover local skills to reference:
```bash
find .claude/skills -name "SKILL.md" 2>/dev/null
ls plugins/*/skills/*/SKILL.md 2>/dev/null
```
Read each skill's frontmatter to understand when to reference it.
## Guideline Keywords
Use these keywords to indicate requirement strength:
- **ALWAYS** — Mandatory requirement
- **NEVER** — Strong prohibition
- **PREFER** — Strong recommendation, exceptions allowed
- **CAN** — Optional, developer's discretion
- **NOTE** — Context or clarification
- **EXAMPLE** — Illustrative example
Strictness hierarchy: ALWAYS/NEVER > PREFER > CAN > NOTE/EXAMPLE
## Writing Rules
- **Existing sections first** - Only propose new sections if no appropriate existing section exists
- **One rule per bullet** - Keep each guideline minimal and atomic
- **Start with keyword** - Every rule begins with ALWAYS/NEVER/PREFER/CAN/NOTE
- **Headers + bullets** - No paragraphs
- **Code blocks** - For commands and templates
- **Reference, don't duplicate** - Point to skills: "See `.claude/skills/db-migrate/SKILL.md`"
- **No filler** - No intros, conclusions, or pleasantries
## Common Sections
Add sections as needed for the project:
### When Stuck
```markdown
## When Stuck
- **ALWAYS** ask a clarifying question or propose alternatives
- **NEVER** initiate large speculative changes without confirmation
```
### Git Commits
```markdown
## Git Commits
- **ALWAYS** write succinct commit messages in imperative mood
- **ALWAYS** keep the first line short
- **NEVER** mention that you are an AI
```
### Issue Management
```markdown
## Linear Issues
- **NEVER** change issue status without explicit instruction
- **NEVER** create issues without explicit instruction
```
### Package Manager
```markdown
## Package Manager
Use **pnpm**: `pnpm install`, `pnpm dev`, `pnpm test`
```
### Local Skills
Reference each discovered skill:
```markdown
## Database
Use `db-migrate` skill. See `.claude/skills/db-migrate/SKILL.md`
```
### Domain-Specific Sections
Add sections for each tech stack (Frontend, Backend, etc.) with domain-specific guidelines.
## Anti-Patterns
Omit these:
- "Welcome to..." or "This document explains..."
- Obvious instructions ("run tests", "write clean code")
- Explanations of why (just say what)
- Long prose paragraphs
- Content duplicated from skills (reference instead)Related Skills
manage-agents
Create, modify, and manage Claude Code subagents with specialized expertise. Use when you need to "work with agents", "create an agent", "modify an agent", "set up a specialist", "I need an agent for [task]", or "agent to handle [domain]". Covers agent file format, YAML frontmatter, system prompts, tool restrictions, MCP integration, model selection, and testing.
langchain-agents
Expert guidance for building LangChain agents with proper tool binding, memory, and configuration. Use when creating agents, configuring models, or setting up tool integrations in LangConfig.
git-commit-for-ai-agents
Commit changes to a git repository. Use whenever a git commit is to be executed.
dispatching-parallel-agents
Use when facing 3+ independent failures that can be investigated without shared state or dependencies. Dispatches multiple Claude agents to investigate and fix independent problems concurrently.
custom-sub-agents
Guidance for creating and organizing custom sub-agents in local repos, including folder conventions, per-agent structure, and AGENTS.md indexing. Use when asked where to store sub-agents or how to document them.
custom-agents
GitHub Custom Agent File Format
creating-agents
Create and review agent definition files (agents.md) that give AI coding agents a clear persona, project knowledge, executable commands, code style examples, and explicit boundaries. Use when a user asks to create an agent, define an agent persona, write an agents.md file, set up a custom Copilot agent, review an existing agent definition, or improve agent quality. Covers the six core areas: commands, testing, project structure, code style, git workflow, and boundaries.
create-agents-md
Create or rewrite AGENTS.md files for Open Mercato packages and modules. Use this skill when adding a new package, creating a new module, or when an existing AGENTS.md needs to be created or refactored. Ensures prescriptive tone, MUST rules, checklists, and consistent structure across all agent guidelines.
building-agents
Expert at creating and modifying Claude Code agents (subagents). Auto-invokes when the user wants to create, update, modify, enhance, validate, or standardize agents, or when modifying agent YAML frontmatter fields (especially 'model', 'tools', 'description'), needs help designing agent architecture, or wants to understand agent capabilities. Also auto-invokes proactively when Claude is about to write agent files (*/agents/*.md), create modular agent architectures, or implement tasks that involve creating agent components.
audit-agents-md
Audit AGENTS.md files for token efficiency, completeness, scope hygiene, and actionability. Also considers skills and Cursor rules for redundancy. Use when the user wants to review, optimize, or restructure project agent instructions.
ai-coding-agents
Comprehensive guide for using Codex CLI (OpenAI) and Claude Code CLI (Anthropic) - AI-powered coding agents. Use when orchestrating CLI commands, automating tasks, configuring agents, or troubleshooting issues.
ai-agents
Production-grade AI agent patterns with MCP integration, agentic RAG, handoff orchestration, multi-layer guardrails, observability, token economics, ROI frameworks, and build-vs-not decision guidance (modern best practices)