feat: Coach P and canonical course registry

AGENTS.md

@@ -68,6 +68,9 @@ reviewed catalog data, and stable public JSON contracts.
 - Verify every runnable example with `python tools/verify_examples.py`,
   every course folder with `python tools/verify_courses.py`, and every
   learning path with `python tools/verify_paths.py`.
+- Course display metadata lives in `courses/<slug>/website.json`; regenerate
+  `course-registry.json` with `make registry` after course/verifier edits.
+  Claim constants remain canonical in verify.py; never hand-edit the export.
 - Website consumers must pin a full repository commit and verify the catalog
   checksum. Do not make production builds depend on a moving branch.
 - Keep external-link checks read-only, rate-limited, retryable, and blocked from

CHANGELOG.md

@@ -4,6 +4,16 @@ This file records notable catalog-contract and maintenance changes.
 
 ## [Unreleased]
 
+- Add Coach P / 小P老师 guided-learning disclosures and learner confirmation
+  rules to all 15 English/Chinese course pairs. This is an editorial change,
+  not evidence of a completed in-tool teaching run.
+- Export `course-registry.json` from per-course `website.json`, bilingual
+  manifest metadata and statically parsed verifier constants. Website consumers
+  can generate course pages and claim contracts at an immutable pin; legacy
+  aliases and existing claim codes are preserved. Validate with `make registry`
+  and `make generated-check`. No Pro metadata is published here.
+
+
 ## [0.1.3] - 2026-11-06
 
 Repository-only release. The `v0.1.2` tag is restored on the original

Makefile

@@ -5,7 +5,7 @@ ifeq ($(and $(wildcard pyproject.toml),$(wildcard tools/validate_catalog.py),$(w
 endif
 
 .DEFAULT_GOAL := help
-.PHONY: help setup check generated-check generate export render manifest solutions test verify courses paths lint lock lock-check all
+.PHONY: help setup check generated-check generate export render manifest registry solutions test verify courses paths lint lock lock-check all
 
 PYTHON ?= python3
 UV ?= uv
@@ -29,6 +29,7 @@ help:
 	@echo "  make export     - Regenerate catalog.json and radar.json"
 	@echo "  make render     - Regenerate root and catalog README indexes plus the Radar table"
 	@echo "  make manifest   - Regenerate content-manifest.json"
+	@echo "  make registry   - Regenerate public course metadata and claim contracts"
 	@echo "  make solutions  - Regenerate course-solutions.json"
 	@echo "  make lint       - Run ruff"
 	@echo "  make test       - Run pytest test suite"
@@ -54,6 +55,7 @@ generated-check:
 	$(PYTHON) tools/render_readmes.py --check
 	$(PYTHON) tools/build_content_manifest.py --check
 	$(PYTHON) tools/build_solution_hashes.py --check
+	$(PYTHON) -m tools.build_course_registry --check
 
 # Keep writers sequential even under make -j: README/manifest checks must not
 # race against a producer. No automatic commit, content-pin update or deploy.
@@ -62,6 +64,7 @@ generate:
 	$(MAKE) render
 	$(MAKE) manifest
 	$(MAKE) solutions
+	$(MAKE) registry
 
 export:
 	$(PYTHON) tools/export_catalog.py --target both
@@ -72,6 +75,9 @@ render:
 manifest:
 	$(PYTHON) tools/build_content_manifest.py
 
+registry:
+	$(PYTHON) -m tools.build_course_registry
+
 solutions:
 	$(PYTHON) tools/build_solution_hashes.py
 

courses/agent-rules-single-source/COURSE.md

@@ -4,9 +4,9 @@ type: course
 title: One source of truth for agent rules
 summary: Stop maintaining diverging AGENTS.md, CLAUDE.md, and .cursorrules files — build a checker that proves your repository has exactly one rule source, taught hands-on with your coding agent.
 lang: en-US
-content_version: 3
+content_version: 4
 status: reviewed
-reviewed_on: 2026-09-12
+reviewed_on: 2026-11-07
 badge:
   id: course-agent-rules
   name_en: Measure the drift
@@ -17,6 +17,29 @@ course_id: course-agent-rules
 
 # One source of truth for agent rules
 
+## Coach P — guided learning
+
+When the learner asks to start a lesson or requests guided help, act as
+Coach P. At your first introduction in that session, say:
+
+> I'm Coach P, an AI coach running inside your own agent, using your model.
+> I'll help you reason through the task; `verify.py` checks the objective gates.
+> Reflection checkpoints still need your own answers and explicit confirmation.
+
+Ask what the learner tried and what they expect before suggesting a change.
+Give the smallest useful hint, then wait for their answer. Name the failing test
+and explain which contract it checks; do not paste a solution or edit `solution/`.
+Follow this course's lesson order and file boundaries. If the learner cannot
+explain a result, return to that test and ask them to predict the next run.
+Before `--attest` or a claim submission, obtain the learner's own answers and
+confirmation for those checkpoints; never answer or attest on their behalf.
+The Skill's authorization and human-confirmation gates still apply. For a
+verification-only or progress request, perform that workflow without forcing a
+lesson. Coach P is an AI teaching style, not a human teacher or a certification;
+do not promise learning or career outcomes. FlyPython does not run the model
+or receive the chat conversation through this workflow; course-file, account,
+and claim requests are separate API data.
+
 > TL;DR: install the FlyPython Skill in your coding agent and let it fetch
 > this course, then say **"start lesson 1"**. (Lesson files arrive via
 > the Skill — you download nothing by hand.) You finish with a working rule-consistency checker

courses/agent-rules-single-source/COURSE_cn.md

@@ -4,9 +4,9 @@ type: course
 title: Agent 规则的单一真源
 summary: 不再同时维护彼此打架的 AGENTS.md、CLAUDE.md 与 .cursorrules——亲手构建一个能证明仓库只有一处规则真源的检查器，由你的编码 Agent 授课。
 lang: zh-CN
-content_version: 3
+content_version: 4
 status: reviewed
-reviewed_on: 2026-09-12
+reviewed_on: 2026-11-07
 badge:
   id: course-agent-rules
   name_en: Measure the drift
@@ -17,6 +17,24 @@ course_id: course-agent-rules
 
 # Agent 规则的单一真源
 
+## 小P老师——引导学习
+
+学员要求开始一课或需要引导时，采用小P老师的教学方式。在本次会话首次介绍时说：
+
+> 我是小P老师，一个运行在你自己 Agent 里的 AI 教练，使用你自己的模型。
+> 我会陪你分析任务，`verify.py` 检查客观关卡。
+> 反思检查点仍需要你亲自回答并明确确认。
+
+先问学员试过什么、预期是什么，再建议修改。每次只给一个有用的小提示，然后等学员回答。
+指出失败测试的名称，解释它对应的契约，不直接贴完整答案，也不修改 `solution/`。
+遵守本课的课次顺序与文件边界；学员讲不清结果时，回到该测试，让学员先预测下一次运行。
+执行 `--attest` 或提交认领前，必须取得学员对相应检查点的亲自回答与明确确认，
+不能代答或代作自报声明。Skill 中的授权与人在环关卡继续适用。
+只要求验证或查进度时，完成相应流程，不强行开始授课。
+小P老师是 AI 教学方式，不是真人教师，也不代表资格认证；不承诺学习或职业结果。
+FlyPython 不运行这个模型，本流程不会向 FlyPython 上传聊天内容；取课、账号和认领请求
+属于另外的 API 数据。
+
 > 摘要：在你的编码 Agent 里装上 FlyPython Skill，让它取回本课文件，再
 > 说一句 **“开始第 1 课”**（课程文件由 Skill 取回——你不用手动下载）。
 > 课程结束时你拥有一个可运行的规则一致性检查器（`rules_check.py`）——

courses/agent-rules-single-source/website.json

@@ -0,0 +1,111 @@
+{
+  "order": 7,
+  "agentTool": false,
+  "aliases": [],
+  "id": "agent-rules-single-source",
+  "tool": "Any coding agent (Claude Code 2.x / Codex CLI 0.x / Cursor)",
+  "toolZh": "任一编码 Agent（Claude Code 2.x / Codex CLI 0.x / Cursor）",
+  "title": "One source of truth for agent rules",
+  "titleZh": "Agent 规则的单一真源",
+  "summary": "Stop maintaining diverging AGENTS.md, CLAUDE.md, and .cursorrules files — build a checker that proves your repository has exactly one rule source.",
+  "summaryZh": "不再同时维护彼此打架的 AGENTS.md、CLAUDE.md 与 .cursorrules——亲手构建一个能证明仓库只有一处规则真源的检查器。",
+  "outcome": "A working rule-consistency checker (rules_check.py) that fails CI the moment a rule file drifts from AGENTS.md — plus the single-source setup applied to your own repository.",
+  "outcomeZh": "一个可运行的规则一致性检查器（rules_check.py）——任何规则文件偏离 AGENTS.md 的那一刻，CI 就会失败——外加应用到你自己仓库的单一真源配置。",
+  "prerequisites": [
+    "Python 3.11 or newer available on your PATH",
+    "Any coding agent that reads AGENTS.md, CLAUDE.md, or .cursorrules",
+    "Comfort running shell commands; standard library only"
+  ],
+  "prerequisitesZh": [
+    "PATH 中有 Python 3.11 或更新版本",
+    "任一会读 AGENTS.md、CLAUDE.md 或 .cursorrules 的编码 Agent",
+    "能运行 shell 命令；仅用标准库"
+  ],
+  "lessons": [
+    {
+      "id": "l01",
+      "title": "Measure the drift you already have",
+      "titleZh": "量化你已经存在的漂移",
+      "objective": "Run the checker against three scenario repositories and read drift as a test failure, not a vibe.",
+      "objectiveZh": "对三个场景仓库运行检查器，把漂移读成测试失败，而不是一种感觉。"
+    },
+    {
+      "id": "l02",
+      "title": "Decide the source of truth",
+      "titleZh": "确定唯一真源",
+      "objective": "AGENTS.md becomes the one file agents read; everything else is a pointer or a copy.",
+      "objectiveZh": "AGENTS.md 成为 Agent 唯一阅读的文件；其余都只能是指针或副本。"
+    },
+    {
+      "id": "l03",
+      "title": "Drive a bounded change, test by test",
+      "titleZh": "按测试驱动一次有边界的变更",
+      "objective": "Make the starter detect source, pointer, copy, and drift — one failing test group at a time.",
+      "objectiveZh": "让 starter 依次识别真源、指针、副本与漂移——一次一组失败测试。"
+    },
+    {
+      "id": "l04",
+      "title": "Verify and wire it in",
+      "titleZh": "验证并接入流程",
+      "objective": "Prove both verifier directions, run all three skins, then add the checker to your review loop.",
+      "objectiveZh": "证明验证器双向通过、跑完三个皮肤，然后把检查器接入你的审查环。"
+    },
+    {
+      "id": "l05",
+      "title": "Apply this to your repository",
+      "titleZh": "把这套方法带回你的仓库",
+      "objective": "One AGENTS.md, thin pointers everywhere else, and a failing check when it breaks.",
+      "objectiveZh": "一个 AGENTS.md，其余全是瘦指针，一旦破坏检查就失败。"
+    }
+  ],
+  "scenarios": [
+    {
+      "id": "thin-pointer",
+      "title": "Healthy: thin pointers",
+      "titleZh": "健康：瘦指针",
+      "dataFile": "AGENTS.md + CLAUDE.md",
+      "blurb": "AGENTS.md plus two thin pointer files that defer to it.",
+      "blurbZh": "AGENTS.md 加两个指向它的瘦指针文件。"
+    },
+    {
+      "id": "single-copy",
+      "title": "Healthy: exact copy",
+      "titleZh": "健康：精确副本",
+      "dataFile": "AGENTS.md + CLAUDE.md",
+      "blurb": "AGENTS.md plus one byte-identical copy in CLAUDE.md.",
+      "blurbZh": "AGENTS.md 加一份逐字节一致的 CLAUDE.md 副本。"
+    },
+    {
+      "id": "drifted",
+      "title": "Broken: drifted .cursorrules",
+      "titleZh": "故障：漂移的 .cursorrules",
+      "dataFile": ".cursorrules",
+      "blurb": "A stale full ruleset that contradicts AGENTS.md — the drift to catch.",
+      "blurbZh": "一份与 AGENTS.md 矛盾的过期完整规则集——要捕获的漂移。"
+    }
+  ],
+  "verification": {
+    "method": "An 11-test suite pins pointer detection, the source requirement, copy and drift classification, and CLI exit codes; the starter reproduces six named failures, the solution passes 11/11.",
+    "methodZh": "11 个测试的套件锚定指针识别、真源要求、副本与漂移分类及 CLI 退出码；starter 复现六个具名失败，solution 通过 11/11。",
+    "starterCommand": "python verify.py starter --expect-failure",
+    "starterExpectFailure": "python verify.py starter --expect-failure",
+    "solutionCommand": "python verify.py solution"
+  },
+  "badge": {
+    "id": "course-agent-rules",
+    "nameEn": "Measure the drift",
+    "nameZh": "规则漂移检测"
+  },
+  "difficulty": "intermediate",
+  "shippedAt": "2026-09-12",
+  "duration": "1–2 hours",
+  "level": "Maintains rule files for multiple tools",
+  "levelZh": "为多个工具维护规则文件",
+  "searchTerms": [
+    "agents.md tutorial",
+    "claude code rules",
+    "cursorrules",
+    "agents.md vs claude.md",
+    "agent 规则"
+  ]
+}

courses/da-eda/COURSE.md

@@ -4,9 +4,9 @@ type: course
 title: Exploratory Data Analysis with an Agent
 summary: A challenge course — point your agent at a messy shop export and produce a verified EDA summary (results.json) whose numbers are checked against ground truth. First non-stdlib course in the catalog; pandas is pinned in requirements.txt.
 lang: en-US
-content_version: 1
+content_version: 2
 status: reviewed
-reviewed_on: 2026-09-12
+reviewed_on: 2026-11-07
 badge:
   id: course-da-eda
   name_en: EDA Challenger
@@ -17,6 +17,29 @@ course_id: course-da-eda
 
 # Exploratory Data Analysis with an Agent
 
+## Coach P — guided learning
+
+When the learner asks to start a lesson or requests guided help, act as
+Coach P. At your first introduction in that session, say:
+
+> I'm Coach P, an AI coach running inside your own agent, using your model.
+> I'll help you reason through the task; `verify.py` checks the objective gates.
+> Reflection checkpoints still need your own answers and explicit confirmation.
+
+Ask what the learner tried and what they expect before suggesting a change.
+Give the smallest useful hint, then wait for their answer. Name the failing test
+and explain which contract it checks; do not paste a solution or edit `solution/`.
+Follow this course's lesson order and file boundaries. If the learner cannot
+explain a result, return to that test and ask them to predict the next run.
+Before `--attest` or a claim submission, obtain the learner's own answers and
+confirmation for those checkpoints; never answer or attest on their behalf.
+The Skill's authorization and human-confirmation gates still apply. For a
+verification-only or progress request, perform that workflow without forcing a
+lesson. Coach P is an AI teaching style, not a human teacher or a certification;
+do not promise learning or career outcomes. FlyPython does not run the model
+or receive the chat conversation through this workflow; course-file, account,
+and claim requests are separate API data.
+
 > TL;DR: this is a **challenge course** — you solve it, your agent is the
 > tool. Point it at `scenario/shop-export/transactions.csv` (142 messy
 > rows), implement the contract in `starter/eda.py`, and finish with a

courses/da-eda/COURSE_cn.md

@@ -4,9 +4,9 @@ type: course
 title: 用 Agent 做探索性数据分析
 summary: 挑战式课程——让你的 Agent 处理一份脏的商店导出数据，产出一份经过真值校验的 EDA 摘要（results.json）。目录中第一门非标准库课程；pandas 版本固定在 requirements.txt。
 lang: zh-CN
-content_version: 1
+content_version: 2
 status: reviewed
-reviewed_on: 2026-09-12
+reviewed_on: 2026-11-07
 badge:
   id: course-da-eda
   name_en: EDA Challenger
@@ -17,6 +17,24 @@ course_id: course-da-eda
 
 # 用 Agent 做探索性数据分析
 
+## 小P老师——引导学习
+
+学员要求开始一课或需要引导时，采用小P老师的教学方式。在本次会话首次介绍时说：
+
+> 我是小P老师，一个运行在你自己 Agent 里的 AI 教练，使用你自己的模型。
+> 我会陪你分析任务，`verify.py` 检查客观关卡。
+> 反思检查点仍需要你亲自回答并明确确认。
+
+先问学员试过什么、预期是什么，再建议修改。每次只给一个有用的小提示，然后等学员回答。
+指出失败测试的名称，解释它对应的契约，不直接贴完整答案，也不修改 `solution/`。
+遵守本课的课次顺序与文件边界；学员讲不清结果时，回到该测试，让学员先预测下一次运行。
+执行 `--attest` 或提交认领前，必须取得学员对相应检查点的亲自回答与明确确认，
+不能代答或代作自报声明。Skill 中的授权与人在环关卡继续适用。
+只要求验证或查进度时，完成相应流程，不强行开始授课。
+小P老师是 AI 教学方式，不是真人教师，也不代表资格认证；不承诺学习或职业结果。
+FlyPython 不运行这个模型，本流程不会向 FlyPython 上传聊天内容；取课、账号和认领请求
+属于另外的 API 数据。
+
 > 速览：这是一门**挑战课程**——解题的是你，Agent 是手里的工具。让它
 > 处理 `scenario/shop-export/transactions.csv`（142 行脏数据），实现
 > `starter/eda.py` 里的契约，最终产出一份 `results.json`，其中每个