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 rules/lawofall/agentcore/intercept-disciplinegit clone --depth 1 https://github.com/Lawofall/AgentCoreWhat 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 | $0.00916 | $0.00916 |
| Opus 5 | $0.00458 | $0.00458 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00092 | $0.00092 |
Grade A, and why
intercept-discipline 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 2d 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.
What it actually says
拦截纪律（硬闸 · 软闸）
产品 AI「不听话」时，禁止默认提案加硬拒 / 软警告 / 扫自由文。拦截极易误伤；软闸也不免费（累计只读软提醒已撤，净负）。
已定案清单与否决表 → 编排器 · 失败与否决；收口姿势 A → 执行引擎 · 可用性诚实性；熔断诚实边界 → 安全权限与治理。本文只留 How（阶梯 / 白名单 / 提案格式）。
阶梯（必须按序；禁止跳级）
- 提示词 / Skill / 结构字段（playbook、
deliverable；禁借机加已删completion_criteriakind 的替代启发式 → 编排器 · 交付验收）。提示词怎么改 →prompt-design.mdc - 观测 + 人审（日志、审批卡、
escalate）——先看见再谈拦 - 一次性软提示（可忽略、不累计、不改成功路径、不扫用户长文猜意图）
- 硬拒——仅同时满足：条件结构化可证明；误伤面可枚举且可接受；不拦会破契约 / 安全底线 / 明确能力缺失
硬闸白名单（仅此类可提案，仍须人确认）
- 结构契约：schema / form / 写盘 scope、合同字段（禁替代启发式）
- 能力缺失：无执行环境却要外环验证 / 长驻就绪等（表述跟工具/契约）
- 灾难安全：熔断级毁灭性动作、敏感读；文案须诚实「启发式兜底，并非完整拦截」。fuse 已覆盖形 → breaker
DENY（禁再提可批可跑 / 仅改文案当终案）→ 安全 · 熔断
默认否决（提了也先当否决项写明）
- 用硬/软闸「优化模型服从度」「防吹牛」「质量启发式」
- 意图分类器：扫用户原文 / task·角色自由文猜意图再分叉（短允许表点名刷新除外）
- 累计计数软提醒、与硬闸同条件「先软再硬」叠床架屋（成篇审计等已有意递进除外）
- 扩大收口/书目/完成话术正则面冒充「近零误报」；禁借「漏拦」再扩面 → 执行引擎
- 边删 kind 边加新启发式完成硬闸；把「验码绿」等伪装成新 kind / 领域 kind 扩表
提建议时的强制格式
若仍要提拦截：单列 误伤面、为何阶梯 1–2 不够、软是否可能净负；标「需人确认」。未写这三项 = 无效提案。与 dev-process.mdc 补丁绊线对齐：拦截类补丁触绊线 → 停、提根因重设计，不直接落地。
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.
- 2d ago First seen · 36 lines · 916 tokens per session scan A 04b5188f08e7
intercept-discipline is a cursor rule published in the GitHub repository Lawofall/AgentCore (78 stars, last pushed 3d ago), licensed MIT. It adds 916 tokens to every session, about $0.0046 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 cursor rules, from other repositories
project-main-rules
You are an expert in building production-ready AI agent applications using Python, FastAPI, LangGraph, and LangChain.
agent-development
Guidelines for developing pydantic-ai agents, tools, and handlers.
part-3-autonomous-research-mode
Autonomous research tiers, brainstorm-to-paper workflow, Tier 3, proof checks, and persistence.
rolemule-core
Core RoleMule conventions — app name, error system, JSON serialization, background tasks.
skills_prompt
Cursor rule "skills_prompt" from nemori-ai/agent_skills, covering what are skills?, when to use skills?, when to create skills?, usage process and tool permissions and file access.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.