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.
npx agentmods add instructions/walterra/claude-code-helpers/claude-mdgit clone --depth 1 https://github.com/walterra/claude-code-helpersWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/walterra/claude-code-helpers/claude-md)<a href="https://agentmods.dev/instructions/walterra/claude-code-helpers/claude-md"><img src="https://agentmods.dev/badge/instructions/walterra/claude-code-helpers/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00719 | $0.00719 |
| Opus 5 | $0.00360 | $0.00360 |
| Sonnet 5 | $0.00144 | $0.00144 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
claude-code-helpers CLAUDE.md scanned grade A with 0 findings 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 4d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Repository Overview
This is a collection of Claude Code helpers and sample commands, containing:
- Custom commands for Claude Code workflow optimization
- Todo implementation system based on badlogic's approach with enhancements
- Git commit helpers with structured workflows
Key Commands and Workflows
Todo Implementation System (commands/todo.md)
A structured workflow system for implementing tasks with isolated work directories and completion tracking. Key principles:
- Tasks are worked on in isolation in
spec/todos/work/[task-name]/directories - Completion tracking happens in
spec/todos/done/ - CRITICAL: Never mark tasks as complete unless ALL success criteria are verifiably met
- NEVER COMMIT OR PUSH: The agent must never run
git commitorgit push- only users handle git operations - Follow the complete Phase 0-4 workflow (Setup → Select → Refine → Implement → Complete)
Critical workflow phases:
- Phase 0 (Setup): Read
spec/todo.mdandspec/project-description.md, handle existing tasks - Phase 1 (Select): Present numbered todos, initialize work folder with agent PID
- Phase 2 (Refine): Define description, success criteria, and implementation plan with user approval
- Phase 3 (Implement): Execute plan with checkboxes, run checks, validate with user tests
- Phase 4 (Complete): Verify ALL criteria met before marking complete
Success criteria must include:
- Functional requirements
- Quality checks (lint, typecheck, tests)
- User validation steps
- Documentation updates
Git Commit Workflow (commands/commit.md)
Structured approach for committing staged files:
- Identify staged files with
git diff --cached --name-only - Commit with one-liner message (60 char max, no semantic prefix, no credit info)
- Handle pre-commit hook issues by running maintenance commands
- NEVER use
git add .or bulk staging - only stage originally staged files - Leave unstaged files (even if modified by formatters) unstaged
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.
- 4d ago First seen · 81 lines · 719 tokens per session scan A 7c31d573bacb
claude-code-helpers CLAUDE.md is an instructions file published in the GitHub repository walterra/claude-code-helpers (5 stars, last pushed 1y ago), licensed Apache-2.0. It adds 719 tokens to every session, about $0.0036 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
agentistics AGENTS.md
AGENTS.md instructions for blpsoares/agentistics, covering before starting implementation work, opening a pr, merging: squash a feature, merge a release, documentation and scope discipline.
event-store-adapter-js AGENTS.md
AGENTS.md instructions for j5ik2o/event-store-adapter-js, covering agents.md, 言語, github pull request のモニタリング, $grill-me 実行時の作業判断 and 軽い作業の場合.
beans CLAUDE.md
Instructions for hmans/beans, covering what we're building, commits, pull requests, project structure and graphql.
roadmap-skill AGENTS.md
Instructions for shiquda/roadmap-skill, covering project guidelines, git commit messages, format, types and examples.
confluence-adf-mcp CLAUDE.md
Claude Code instructions for karbassi/confluence-adf-mcp, covering project instructions and sdlc.
Lucerne CLAUDE.md
Claude Code instructions for L-K-M/Lucerne, covering pull request babysitting, respond to every round on its merits and declare steady-state and stop when any of these hold.