agentmods.dev

MiHaCoBench: Instructions file for Claude Code

CLAUDE.md

MiHaCoBench CLAUDE.md is an instructions file for Claude Code from HangYu8123/MiHaCoBench. It costs 1,141 tokens per session, scanned A, a copy of MiHaCoBench AGENTS.md, MIT.

A rulebook for running Claude Code workflows in the MiHaCoBench repository. It explains how Claude Code, a coding assistant, finds instructions, resolves file paths, and decides whether a prompt is allowed to start a workflow.

In plain words
What is it for?
Use it when executing or modifying MiHaCoBench workflows, resolving installed-versus-source paths, or checking repository context and workflow rules.
Why use it?
It prevents workflows from starting on incomplete or incorrectly formatted requests and helps the assistant use the right repository files.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions subagents; mentions Claude Code.

This is HangYu8123/MiHaCoBench's own configuration. It tells Claude Code how to work on MiHaCoBench itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything MiHaCoBench configures →

not rated 44repo 3mo ago A scan 1,141 tokens copy · 86% MIT
View source ↗ HangYu8123/MiHaCoBench
Reuse

Borrowing it

Nothing to install: this file belongs to HangYu8123/MiHaCoBench. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/HangYu8123/MiHaCoBench/master/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/HangYu8123/MiHaCoBench

Made for: Claude Code.

Wrote 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.

agentmods badge for MiHaCoBench CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/hangyu8123/mihacobench/claude-md/github.svg)](https://agentmods.dev/instructions/hangyu8123/mihacobench/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/hangyu8123/mihacobench/claude-md"><img src="https://agentmods.dev/badge/instructions/hangyu8123/mihacobench/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for MiHaCoBench CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/hangyu8123/mihacobench/claude-md"><img src="https://agentmods.dev/badge/instructions/hangyu8123/mihacobench/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,141 This file is loaded in full into every session.
When invoked 1,141 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 86% copy Near-identical to another mod 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01141 $0.01141
Opus 5.5 $0.00456 $0.00456
Sonnet 5.5 $0.00228 $0.00228
Haiku 4.5 $0.00114 $0.00114

Measured 2mo ago against content hash 16a0c2ff7fa9, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

MiHaCoBench 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 2mo 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.

Origin

This is a copy

86% identical to MiHaCoBench AGENTS.md — 13 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

CLAUDE.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — Claude Code CLI Entry Point

This file is auto-discovered by Claude Code when run from the repo root. It is the Claude Code CLI equivalent of copilot-instructions.md.

Pack Path Resolution

Resolve all pack-relative paths in this order:

  1. .github/HarnessFlow/<path> from the target repo root (installed layout).
  2. <path> from the repo root (source repo / pack root layout).

Apply this rule to every path referenced in this file, workflow files, and agent definitions.

Workflow Execution

Step 0 — gate check, always first. Only a filled-in prompt from request_template/ (a mode: header block plus a numbered "Hard constraints" list naming its category's *.instructions.md) triggers a workflow. Any other prompt — even one that mentions workflows or pack files — fails the gate: skip steps 1–5 and answer it normally, with no classification, no *.instructions.md, and no contract/philosophy reads as workflow setup (the Engineering Guidelines below still apply).

For a prompt that passes the gate:

  1. Read and follow, as the main agent and in their entirety, C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/_lib/workflow_contract.md and C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/philosophy/philosophy.instructions.md (resolved via Pack Path Resolution). Do this before any workflow-specific work. Never delegate this read to a subagent in place of doing it yourself.
  2. Read the matched instruction file in its entirety.
  3. Require every subagent to read and follow C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/_lib/workflow_contract.md and C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/philosophy/philosophy.instructions.md (resolved via Pack Path Resolution) before doing workflow-specific work.
  4. Subagent model: Create every subagent on the model the instructions specify — the subagent_model header (see C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/_lib/workflow_contract.md §Subagent Launch Contract). In Claude Code the main agent sets the subagent's model when spawning it: a specific subagent_model id runs the subagent on that exact id (a deliberate override — honor it even if smaller), while inherit or unset uses the main agent's model with no downgrade. On other platforms, follow the Subagent Launch Contract's model-selection steps in C:/Users/Hang Yu/Desktop/MiHaCoBench/.github/HarnessFlow/_lib/workflow_contract.md.
  5. Follow the matched instruction file step-by-step to complete the request.

Read the full file on GitHub · 53 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.

  1. 2mo ago First seen · 53 lines · 1,141 tokens per session scan A 16a0c2ff7fa9

Subscribe to this mod's changes

MiHaCoBench CLAUDE.md is an instructions file published in the GitHub repository HangYu8123/MiHaCoBench (44 stars, last pushed 3mo ago), licensed MIT. It adds 1,141 tokens to every session, about $0.0046 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 86% identical to MiHaCoBench AGENTS.md, differing in 13 lines, and is treated as a copy.

Related

MiHaCoBench CLAUDE.md compared with instructions from other repositories

MiHaCoBench CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
MiHaCoBench CLAUDE.mdHangYu8123/MiHaCoBenchInstructions file443mo agoMITA1,141Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,516GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA2,022GitHub Copilot