What is agent-instructions CLAUDE.md?

Repository instructions for a system that generates slash commands for AI coding agents from reusable text fragments.

Use them to edit command sources or fragments, handle optional features such as beads, generate local commands, and maintain committed build outputs.

They explain how to keep shared command content in one source and prevent generated commands from drifting apart.

agent-instructions CLAUDE.md — Instructions file by wbern

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/wbern/agent-instructions/claude-md

Clone the repo

git clone --depth 1 https://github.com/wbern/agent-instructions

Per session 1,813 This file is loaded in full into every session.

When invoked 1,813 The same file — it is already loaded in full.

Security scan B 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.01813	$0.01813
Opus 5	$0.00907	$0.00907
Sonnet 5	$0.00363	$0.00363
Haiku 4.5	$0.00181	$0.00181

Measured 3d ago against content hash 9a9765e658f6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

agent-instructions CLAUDE.md scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Unrestricted tool accessmediumExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

- **Never** use wildcard patterns like `Bash(git:*)` or `allowed-tools: *`

CLAUDE.md · 202 lines

How it starts

The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Instructions Repository

Note: This file contains instructions for managing THIS repository. Users installing the commands should NOT copy this file - it's for maintainers only.

GitHub URL: https://github.com/wbern/agent-instructions (NOT KenDev-AB)

This repository contains a fragment-based system for generating slash commands for AI coding agents (Claude Code, OpenCode) with TDD focus.

Architecture

Fragment System

Sources (src/sources/*.md): Command templates with INCLUDE directives only
Fragments (src/fragments/*.md): Reusable content blocks
Local commands (.claude/commands/, .opencode/commands/): Generated commands for local development
Dynamic generation: CLI generates commands on-the-fly with optional feature flags (e.g., beads)

Key Principles

Single source of truth: Content lives in fragments, not sources
Clean sources: Source files contain only INCLUDE directives, never expanded content
Track artifacts: Both sources and built commands are committed for easy distribution
Feature flags: Use featureFlag='beads' for conditional content

Contributor Commands

Source files prefixed with underscore (_*.md) are "contributor commands":

Excluded from npm package distribution (consumers never see them)
Included in this repo's local commands for maintainers
Underscore prefix is stripped from output filename (_foo.md → foo.md)
Not listed in README command list
Use --include-contrib-commands internal flag to include them

Development Workflow

Package Manager

Use pnpm for all package operations

Common Commands

pnpm build          # Build README.md and local commands
pnpm test           # Run tests
pnpm vitest run -u  # Update snapshots

Git Hooks (Husky)

Pre-commit hook automatically:

Builds (includes markdownlint --fix)
Runs tests
Stages generated files (README.md, generated commands)

This ensures artifacts are always in sync with sources.

Read the full file on GitHub · 202 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago First seen · 202 lines · 1,813 tokens per session scan B 9a9765e658f6

Subscribe to this mod's changes

agent-instructions CLAUDE.md is an instructions file published in the GitHub repository wbern/agent-instructions (168 stars, last pushed 3mo ago), licensed MIT. It adds 1,813 tokens to every session, about $0.0091 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

agent-instructions CLAUDE.md