dd1de0d1e9
架构转变：从代运营多客户模式改为开源单用户模式。 - 去掉 clients/ 目录，style.yaml/history.yaml 扁平化到 skill root - Step 1 简化（不再提取客户名，直接读 style.yaml） - 新增 Step 0 环境检查（config/依赖/API 配置，降级标记传递到后续 Step） - Onboard 改为首次设置流程（交互式问答 + 支持"用默认的直接写"） - 3 个脚本去掉 --client 参数，路径扁平化 - 修复 10 项 workflow 问题（降级传递、历史写入、wechat-constraints 引用等） - evals 更新为单用户模式的 3 个场景 - 新增 style.example.yaml 作为默认模板 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
274 lines
8.7 KiB
Python
274 lines
8.7 KiB
Python
#!/usr/bin/env python3
|
|
"""
|
|
Learn from human edits by diffing AI draft vs published final.
|
|
|
|
Compares the original AI-generated article with the human-edited version,
|
|
categorizes the changes, and saves lessons to lessons/.
|
|
|
|
When 5+ lessons accumulate, outputs a prompt for the Agent to update playbook.md.
|
|
|
|
Usage:
|
|
python3 learn_edits.py --draft path/to/draft.md --final path/to/final.md
|
|
python3 learn_edits.py --summarize # summarize all lessons
|
|
|
|
The script does structural analysis; the Agent (LLM) interprets the diffs
|
|
and writes the lesson YAML + playbook updates.
|
|
"""
|
|
|
|
import argparse
|
|
import difflib
|
|
import json
|
|
import re
|
|
import sys
|
|
from datetime import datetime
|
|
from pathlib import Path
|
|
|
|
import yaml
|
|
|
|
SKILL_DIR = Path(__file__).parent.parent
|
|
|
|
|
|
def load_text(path: str) -> str:
|
|
return Path(path).read_text(encoding="utf-8")
|
|
|
|
|
|
def split_sections(text: str) -> list[dict]:
|
|
"""Split markdown into sections by H2 headers."""
|
|
sections = []
|
|
current = {"header": "(intro)", "lines": []}
|
|
|
|
for line in text.split("\n"):
|
|
if line.strip().startswith("## "):
|
|
if current["lines"] or current["header"] != "(intro)":
|
|
sections.append(current)
|
|
current = {"header": line.strip(), "lines": []}
|
|
else:
|
|
current["lines"].append(line)
|
|
|
|
sections.append(current)
|
|
return sections
|
|
|
|
|
|
def extract_title(text: str) -> str:
|
|
for line in text.split("\n"):
|
|
if line.strip().startswith("# ") and not line.strip().startswith("## "):
|
|
return line.strip()[2:].strip()
|
|
return ""
|
|
|
|
|
|
def compute_diff(draft: str, final: str) -> dict:
|
|
"""Compute structured diff between draft and final."""
|
|
draft_lines = draft.split("\n")
|
|
final_lines = final.split("\n")
|
|
|
|
# Line-level diff
|
|
differ = difflib.unified_diff(draft_lines, final_lines, lineterm="")
|
|
diff_lines = list(differ)
|
|
|
|
# Categorize changes
|
|
additions = []
|
|
deletions = []
|
|
for line in diff_lines:
|
|
if line.startswith("+") and not line.startswith("+++"):
|
|
additions.append(line[1:].strip())
|
|
elif line.startswith("-") and not line.startswith("---"):
|
|
deletions.append(line[1:].strip())
|
|
|
|
# Filter empty lines
|
|
additions = [l for l in additions if l]
|
|
deletions = [l for l in deletions if l]
|
|
|
|
# Title change
|
|
draft_title = extract_title(draft)
|
|
final_title = extract_title(final)
|
|
title_changed = draft_title != final_title
|
|
|
|
# Section-level analysis
|
|
draft_sections = split_sections(draft)
|
|
final_sections = split_sections(final)
|
|
draft_h2s = [s["header"] for s in draft_sections if s["header"] != "(intro)"]
|
|
final_h2s = [s["header"] for s in final_sections if s["header"] != "(intro)"]
|
|
structure_changed = draft_h2s != final_h2s
|
|
|
|
# Word count change
|
|
draft_chars = len(draft.replace("\n", "").replace(" ", ""))
|
|
final_chars = len(final.replace("\n", "").replace(" ", ""))
|
|
|
|
return {
|
|
"title_changed": title_changed,
|
|
"draft_title": draft_title,
|
|
"final_title": final_title,
|
|
"structure_changed": structure_changed,
|
|
"draft_h2s": draft_h2s,
|
|
"final_h2s": final_h2s,
|
|
"lines_added": len(additions),
|
|
"lines_deleted": len(deletions),
|
|
"draft_chars": draft_chars,
|
|
"final_chars": final_chars,
|
|
"char_diff": final_chars - draft_chars,
|
|
"additions_sample": additions[:20],
|
|
"deletions_sample": deletions[:20],
|
|
}
|
|
|
|
|
|
def save_diff_for_analysis(diff_result: dict, draft_path: str, final_path: str):
|
|
"""Save diff data for Agent to analyze and write lessons."""
|
|
lessons_dir = SKILL_DIR / "lessons"
|
|
lessons_dir.mkdir(parents=True, exist_ok=True)
|
|
|
|
date_str = datetime.now().strftime("%Y-%m-%d")
|
|
diff_file = lessons_dir / f"{date_str}-diff.yaml"
|
|
|
|
# If file exists, append a counter
|
|
counter = 1
|
|
while diff_file.exists():
|
|
diff_file = lessons_dir / f"{date_str}-diff-{counter}.yaml"
|
|
counter += 1
|
|
|
|
data = {
|
|
"date": date_str,
|
|
"draft_file": str(draft_path),
|
|
"final_file": str(final_path),
|
|
"diff_summary": {
|
|
"title_changed": diff_result["title_changed"],
|
|
"draft_title": diff_result["draft_title"],
|
|
"final_title": diff_result["final_title"],
|
|
"structure_changed": diff_result["structure_changed"],
|
|
"lines_added": diff_result["lines_added"],
|
|
"lines_deleted": diff_result["lines_deleted"],
|
|
"char_diff": diff_result["char_diff"],
|
|
},
|
|
"edits": [], # Agent fills this after analysis
|
|
"patterns": [], # Agent fills this after analysis
|
|
}
|
|
|
|
with open(diff_file, "w", encoding="utf-8") as f:
|
|
yaml.dump(data, f, allow_unicode=True, default_flow_style=False)
|
|
|
|
return diff_file
|
|
|
|
|
|
def count_lessons() -> int:
|
|
"""Count existing lesson files."""
|
|
lessons_dir = SKILL_DIR / "lessons"
|
|
if not lessons_dir.exists():
|
|
return 0
|
|
return len(list(lessons_dir.glob("*-diff*.yaml")))
|
|
|
|
|
|
def summarize_lessons():
|
|
"""Load all lessons and output for Agent to update playbook."""
|
|
lessons_dir = SKILL_DIR / "lessons"
|
|
if not lessons_dir.exists():
|
|
print("No lessons directory found.")
|
|
return
|
|
|
|
lesson_files = sorted(lessons_dir.glob("*-diff*.yaml"))
|
|
if not lesson_files:
|
|
print("No lessons found.")
|
|
return
|
|
|
|
all_lessons = []
|
|
for f in lesson_files:
|
|
with open(f, "r", encoding="utf-8") as fh:
|
|
data = yaml.safe_load(fh)
|
|
if data:
|
|
all_lessons.append(data)
|
|
|
|
print(f"Total lessons: {len(all_lessons)}")
|
|
print(json.dumps(all_lessons, ensure_ascii=False, indent=2))
|
|
|
|
|
|
def main():
|
|
parser = argparse.ArgumentParser(description="Learn from human edits")
|
|
parser.add_argument("--draft", help="Path to AI draft")
|
|
parser.add_argument("--final", help="Path to human-edited final")
|
|
parser.add_argument("--summarize", action="store_true", help="Summarize all lessons")
|
|
args = parser.parse_args()
|
|
|
|
if args.summarize:
|
|
summarize_lessons()
|
|
return
|
|
|
|
if not args.draft or not args.final:
|
|
print("Error: --draft and --final required", file=sys.stderr)
|
|
sys.exit(1)
|
|
|
|
# Load texts
|
|
draft = load_text(args.draft)
|
|
final = load_text(args.final)
|
|
|
|
# Compute diff
|
|
diff_result = compute_diff(draft, final)
|
|
|
|
# Print summary
|
|
print("=" * 60)
|
|
print("EDIT ANALYSIS")
|
|
print("=" * 60)
|
|
|
|
if diff_result["title_changed"]:
|
|
print(f"\n标题修改:")
|
|
print(f" AI: {diff_result['draft_title']}")
|
|
print(f" 人工: {diff_result['final_title']}")
|
|
|
|
if diff_result["structure_changed"]:
|
|
print(f"\n结构修改:")
|
|
print(f" AI H2: {diff_result['draft_h2s']}")
|
|
print(f" 人工 H2: {diff_result['final_h2s']}")
|
|
|
|
print(f"\n数量变化:")
|
|
print(f" 新增 {diff_result['lines_added']} 行, 删除 {diff_result['lines_deleted']} 行")
|
|
print(f" 字数变化: {diff_result['char_diff']:+d} ({diff_result['draft_chars']} → {diff_result['final_chars']})")
|
|
|
|
if diff_result["deletions_sample"]:
|
|
print(f"\n被删除的内容（采样）:")
|
|
for line in diff_result["deletions_sample"][:10]:
|
|
print(f" - {line[:80]}")
|
|
|
|
if diff_result["additions_sample"]:
|
|
print(f"\n新增的内容（采样）:")
|
|
for line in diff_result["additions_sample"][:10]:
|
|
print(f" + {line[:80]}")
|
|
|
|
# Save for Agent analysis
|
|
diff_file = save_diff_for_analysis(diff_result, args.draft, args.final)
|
|
print(f"\nDiff saved to: {diff_file}")
|
|
|
|
# Check if playbook update should be triggered
|
|
lesson_count = count_lessons()
|
|
print(f"Total lessons: {lesson_count}")
|
|
|
|
if lesson_count >= 5 and lesson_count % 5 == 0:
|
|
print(f"\n{'='*60}")
|
|
print("PLAYBOOK UPDATE TRIGGERED")
|
|
print(f"{'='*60}")
|
|
print(f"{lesson_count} lessons accumulated. Agent should:")
|
|
print(f"1. Read all lessons: python3 learn_edits.py --summarize")
|
|
print(f"2. Read current playbook: playbook.md")
|
|
print(f"3. Update playbook with recurring patterns from lessons")
|
|
|
|
# Output instructions for Agent
|
|
print(f"""
|
|
{'='*60}
|
|
INSTRUCTIONS FOR AGENT
|
|
{'='*60}
|
|
|
|
Read the draft and final versions, then analyze the edits:
|
|
|
|
1. Read: {args.draft}
|
|
2. Read: {args.final}
|
|
3. For each meaningful edit, classify it:
|
|
- type: "用词替换" / "段落删除" / "段落新增" / "结构调整" / "标题修改" / "语气调整"
|
|
- before: (original text)
|
|
- after: (edited text)
|
|
- pattern: (what this tells us about the user's preference)
|
|
|
|
4. Update {diff_file} with the edits and patterns lists.
|
|
|
|
5. If this is a recurring pattern (seen in previous lessons too),
|
|
consider updating playbook.md.
|
|
""")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|