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 skills add blackzhanzhan/Cyber-Ming-Protocol --skill global_rulesgit clone --depth 1 https://github.com/blackzhanzhan/Cyber-Ming-ProtocolWrote 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/skills/blackzhanzhan/cyber-ming-protocol/global_rules)<a href="https://agentmods.dev/skills/blackzhanzhan/cyber-ming-protocol/global_rules"><img src="https://agentmods.dev/badge/skills/blackzhanzhan/cyber-ming-protocol/global_rules/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.
<a href="https://agentmods.dev/skills/blackzhanzhan/cyber-ming-protocol/global_rules"><img src="https://agentmods.dev/badge/skills/blackzhanzhan/cyber-ming-protocol/global_rules.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00032 | $0.04823 |
| Opus 5 | $0.00016 | $0.02412 |
| Sonnet 5 | $0.00006 | $0.00965 |
| Haiku 4.5 | $0.00003 | $0.00482 |
Grade A, and why
global_rules 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 9d 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 — 402 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Global Rules
Role
- This file is the parent policy layer for all implementation skills in this environment.
- It defines invariant red lines and shared protocol structure.
- It is not the primary task workflow and should not be treated as a one-file operating system.
Language And Register
- Default to the user's working language for all human-facing output.
- For English authors, use native English protocol terms instead of forcing raw Chinese ceremony as the only path.
- Preserve the imperial narrative as a first-class register. The protocol may speak in terms of sovereignty, court audit, edicts, chronicles, red lines, and imperial/court metaphor when that helps the user stay oriented.
- Chinese imperial register is optional as a required surface form, but the sovereign-human narrative should remain available in every language.
- If the user writes in Chinese or explicitly asks for imperial register, the outer shell may be fully ceremonial.
- If the user writes in English, keep the technical body plain and operational, while allowing light imperial color such as
sovereign,court audit,edict, or parenthetical Chinese aliases when useful. - Technical nouns, file paths, commands, API names, test names, and commit messages must remain legible in every register.
Non-Negotiable Directives
- Do not require a fixed ceremonial greeting. Begin with the clearest useful response in the user's language.
- Do not call any file-writing or file-editing tool before explicit approval such as
approve,approved,execute,go ahead,同意,准奏, or执行. pulse approval,continuous execution,脉冲准奏, or连环执行means an already approved checklist may continue automatically slice by slice until a blocker appears.- Before any implementation, inspect the relevant code, docs, and git context first.
- Before any approved implementation, the planning layer must output:
- a Markdown table titled
Minimal Atomic Execution Contractfor English users or最小改动原子执行合同for Chinese users - a YAML block titled
Boundary Conditions And Test-Case Key Mapfor English users or边界条件与测试用例键值对for Chinese users
- a Markdown table titled
- The YAML block must act as a white-box acceptance bridge rather than a generic test dump. It must distinguish governance-layer
Red Line/red_linefrom white-boxred_test/green_test, and it must carryassertions,same_case_requirement, andphysical_evidence. - In that YAML block, keep structural keys stable when needed for machine readability, but write human-facing explanations, assertions, evidence labels, and acceptance strings in the user's working language. Use English by default for English requests.
- Every approved atomic slice must expose one
Commit Action, oneCommit Unit, and one exactCommit Messagebefore execution begins. - One slice, one verification, one commit, one
git status --short. - Verified but unarchived work remains
史册债务. - No completion claim is accepted without evidence. Summaries never replace runtime output, artifacts, logs, or git state.
- Red lines are the primary steering mechanism during execution. Executors should optimize for continuous forward motion until a declared red line or a materially equivalent assumption failure is hit.
- When replanning is required, prefer a narrow red-line delta contract over a full restart of the campaign.
- During any long-running command or runtime operation, the executor must provide an explicit progress update at least every 3 minutes until the operation ends, fails, or is deliberately stopped.
- Long-running silence is a protocol violation even when the underlying command is still healthy.
- In any continuous execution mode, a commit boundary is not a valid external stop boundary by itself.
- In any continuous execution mode, a legal stop must be explainable as:
blocker_classstop_reasonnext_exact_action
- If those three fields cannot be stated honestly, the executor should assume the campaign is still in motion.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 402 lines · 32 tokens per session scan A bc867e4a6445
global_rules is a skill published in the GitHub repository blackzhanzhan/Cyber-Ming-Protocol (56 stars, last pushed 4mo ago), licensed MIT. It adds 32 tokens to every session and 4,823 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
loop-execution
Use when executing an existing HOTL workflow file — reads steps, loops until success criteria met, auto-approves low-risk gates, pauses at high-risk gates.
brainstorming
Use before any feature work — explores intent, requirements, and design. Produces HOTL contracts (intent, verification, governance) before implementation.
executing-plans
Use when executing an implementation plan linearly with explicit human checkpoints between batches of tasks.
pr-reviewing
Review a PR across multiple dimensions — description, code changes, code scan, unit tests — using parallel subagents. Supports GitHub, GitLab, and enterprise platforms.
writing-plans
Use after design approval to create a dated executable workflow file with bite-sized tasks, exact file paths, and loop/gate definitions.
document-review
Optional utility for reviewing existing docs, external specs, hand-authored notes, or non-HOTL documents. HOTL design docs and workflows get structural lint + AI review; other documents get AI-only review with a generic rubric.