Thoth Pro
/install thoth-pro
Thoth Pro -- Full Auto-Documentation Suite
The most complete documentation tool on ClawHub. Reads your entire codebase, generates professional docs, injects missing docstrings, and builds a CHANGELOG from your git history.
Bundle deal: All 5 Egyptian skills for $29 -> ko-fi.com/s/7625accf3f (save $16)
Step 1 -- Install
pip3 install rich --break-system-packages --quiet
Step 2 -- Validate and deep scan
import os, sys, ast, subprocess, re
from datetime import datetime
from rich.console import Console
from rich.panel import Panel
from rich.table import Table
from rich.rule import Rule
from rich import box
console = Console()
LICENSE_KEY = os.environ.get("LICENSE_KEY", "").strip()
if not LICENSE_KEY or not LICENSE_KEY.startswith("THOTH-PRO-"):
console.print(Panel(
"[red bold]Thoth Pro requires a license key.[/red bold]\
\
"
"Get your key at: [bold cyan]ko-fi.com/occupythemilkyway[/bold cyan]\
\
"
"Or try Thoth Standard ($5): [dim]openclaw skills install thoth[/dim]",
title="License Required",
border_style="red"
))
sys.exit(1)
PROJECT_PATH = os.environ.get("PROJECT_PATH", "").strip()
PROJECT_NAME = os.environ.get("PROJECT_NAME", "").strip()
OUTPUT_DIR = os.environ.get("OUTPUT_DIR", "./docs").strip()
INJECT_DOCSTRINGS = os.environ.get("INJECT_DOCSTRINGS", "yes").lower().strip() == "yes"
GEN_CHANGELOG = os.environ.get("GENERATE_CHANGELOG", "yes").lower().strip() == "yes"
if not PROJECT_PATH or not os.path.exists(PROJECT_PATH):
console.print(Panel(f"[red]PROJECT_PATH not found: {PROJECT_PATH}[/red]", title="Error", border_style="red"))
sys.exit(1)
if not PROJECT_NAME:
PROJECT_NAME = os.path.basename(os.path.abspath(PROJECT_PATH))
os.makedirs(OUTPUT_DIR, exist_ok=True)
console.print()
console.print(Panel.fit(
f"[bold cyan]📜 Thoth Pro -- Full Documentation Suite[/bold cyan]\
"
f"Project: [yellow]{PROJECT_NAME}[/yellow]\
"
f"Inject docstrings: [white]{'Yes' if INJECT_DOCSTRINGS else 'No'}[/white]\
"
f"Git changelog: [white]{'Yes' if GEN_CHANGELOG else 'No'}[/white]\
"
f"Output: [green]{OUTPUT_DIR}/[/green]",
border_style="cyan"
))
SKIP_DIRS = {".git", "__pycache__", "node_modules", ".venv", "venv", "dist", "build", ".next", ".pytest_cache"}
CODE_EXTS = {".py", ".js", ".ts", ".jsx", ".tsx", ".go", ".rs", ".java", ".rb", ".cs", ".cpp", ".c", ".sh"}
META_FILES = {"requirements.txt", "package.json", "pyproject.toml", "go.mod", "Cargo.toml", "setup.py"}
all_files, meta_files, py_symbols, missing_docstrings = [], {}, [], []
for root, dirs, fnames in os.walk(PROJECT_PATH):
dirs[:] = [d for d in dirs if d not in SKIP_DIRS]
for fname in fnames:
full = os.path.join(root, fname)
rel = os.path.relpath(full, PROJECT_PATH)
ext = os.path.splitext(fname)[1].lower()
if fname in META_FILES:
try:
with open(full, encoding="utf-8", errors="replace") as fh:
meta_files[fname] = fh.read(2000)
except Exception:
pass
if ext in CODE_EXTS:
all_files.append((rel, full, ext))
# Deep Python analysis
for rel, full, ext in all_files:
if ext == ".py":
try:
with open(full, encoding="utf-8", errors="replace") as fh:
source = fh.read()
tree = ast.parse(source)
for node in ast.walk(tree):
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef, ast.ClassDef)):
doc = ast.get_docstring(node) or ""
args = []
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)):
args = [a.arg for a in node.args.args if a.arg != "self"]
sym = {"file": rel, "name": node.name, "args": args,
"type": "class" if isinstance(node, ast.ClassDef) else "function",
"doc": doc, "line": node.lineno}
py_symbols.append(sym)
if not doc and not node.name.startswith("_"):
missing_docstrings.append(sym)
except Exception:
pass
# Git changelog
git_log = ""
if GEN_CHANGELOG:
try:
result = subprocess.run(
["git", "-C", PROJECT_PATH, "log", "--oneline", "--no-merges", "-50"],
capture_output=True, text=True, timeout=10
)
git_log = result.stdout.strip()
except Exception:
git_log = ""
tbl = Table(title=f"Deep Scan: {PROJECT_NAME}", box=box.ROUNDED, border_style="cyan")
tbl.add_column("Metric", style="dim")
tbl.add_column("Value", style="yellow")
tbl.add_row("Code files", str(len(all_files)))
tbl.add_row("Python symbols", str(len(py_symbols)))
tbl.add_row("Missing docstrings", str(len(missing_docstrings)))
tbl.add_row("Git commits found", str(len(git_log.splitlines())) if git_log else "No git history")
tbl.add_row("Output directory", OUTPUT_DIR)
console.print(tbl)
console.print(Rule("[cyan]File contents[/cyan]"))
for rel, full, ext in sorted(all_files)[:30]:
try:
with open(full, encoding="utf-8", errors="replace") as fh:
content = fh.read(5000)
print(f"\
=== {rel} ===\
{content}\
=== END {rel} ===")
except Exception:
pass
if git_log:
console.print(Rule("[cyan]Git history[/cyan]"))
print(git_log)
console.print(Rule("[cyan]Symbols missing docstrings[/cyan]"))
for sym in missing_docstrings[:40]:
args_str = ", ".join(sym["args"])
print(f" {sym['file']}:{sym['line']} -- {sym['type']} {sym['name']}({args_str})")
Step 3 -- Generate full documentation suite
Generate five documents based on everything scanned above:
1. README.md -- Complete project overview, install, usage, examples, config, structure, license 2. API_REFERENCE.md -- Every public function/class/method with parameters, return types, description, example 3. USAGE_GUIDE.md -- Practical guide with real-world worked examples and common patterns 4. ARCHITECTURE.md -- How the codebase is structured, key design decisions, data flow, module dependencies 5. CHANGELOG.md -- If git history was provided, group commits into version sections (Features / Fixes / Refactors)
For INJECT_DOCSTRINGS=yes: Also produce a JSON mapping of {"file": "rel_path", "line": N, "docstring": "..."} for every symbol in the missing_docstrings list. Claude will use this to inject docstrings.
Write everything with the depth and polish of a senior technical writer.
Step 4 -- Save everything
import os, json
from rich.console import Console
from rich.panel import Panel
from rich.table import Table
from rich import box
console = Console()
# (Claude saves all files to OUTPUT_DIR)
files_written = [
(os.path.join(OUTPUT_DIR, "README.md"), "Project overview"),
(os.path.join(OUTPUT_DIR, "API_REFERENCE.md"), "Full API reference"),
(os.path.join(OUTPUT_DIR, "USAGE_GUIDE.md"), "Usage examples"),
(os.path.join(OUTPUT_DIR, "ARCHITECTURE.md"), "Architecture overview"),
(os.path.join(OUTPUT_DIR, "CHANGELOG.md"), "Version changelog"),
]
tbl = Table(title="Thoth Pro -- Documentation Complete", box=box.SIMPLE, border_style="green")
tbl.add_column("File", style="cyan")
tbl.add_column("Description", style="dim")
for path, desc in files_written:
tbl.add_row(path, desc)
console.print()
console.print(tbl)
console.print(Panel(
"[bold green]Full documentation suite generated.[/bold green]\
\
"
f"[yellow]{len(files_written)} documents[/yellow] written to [cyan]{OUTPUT_DIR}[/cyan]\
"
f"Symbols documented: [yellow]{len(py_symbols)}[/yellow]",
border_style="green"
))
After generating all documents in Step 3, save each one to OUTPUT_DIR using your file writing tool.
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install thoth-pro - 安装完成后,直接呼叫该 Skill 的名称或使用
/thoth-pro触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
Thoth Pro 是什么?
Thoth Pro -- Full Auto-Documentation Suite. README + API reference + docstrings injected into source + CHANGELOG from git + architecture doc. The definitive... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 42 次。
如何安装 Thoth Pro?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install thoth-pro」即可一键安装,无需额外配置。
Thoth Pro 是免费的吗?
是的,Thoth Pro 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Thoth Pro 支持哪些平台?
Thoth Pro 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Thoth Pro?
由 OccupyTheMilkyWay(@occupythemilkyway)开发并维护,当前版本 v1.0.0。