Dianel555/DSkills / skills/agent-wiki
agent-wiki
UnverifiedIncremental LLM-friendly wiki generator for Obsidian note vaults. Use when: (1) Building wiki from notes, (2) Ingesting notes to wiki, (3) Obsidian LLM wiki, (4) Incremental knowledge base management. Triggers: 'build wiki from notes', 'ingest notes to wiki', 'Obsidian LLM wiki', 'incremental knowledge base'.
scripts/agent_wiki/scanner.py
View on GitHub"""Source file scanning and classification."""
from __future__ import annotations
import fnmatch
from collections.abc import Callable, Iterator
from pathlib import Path
from typing import Any
from . import boundary, config
from . import cache as cache_mod
EXCLUDED_DIRS = {"wiki", ".obsidian", "attachments", ".git", ".trash"}
def _ignore_patterns(vault: Path, warnings: list[dict[str, str]]) -> list[str]:
ignore = vault / ".wikiignore"
try:
# The existence probe is inside the same protection as the read: an
# inspection failure (e.g. access denied) must degrade to "no ignore rules
# applied" plus a warning, not abort the whole classification.
if not boundary.exists(ignore):
return []
text = boundary.read_text(ignore, encoding="utf-8")
except (OSError, UnicodeDecodeError, boundary.BoundaryError):
# Unreadable ignore rules must not silently change what gets scanned: report
# the failure so the caller can decide, instead of scanning with default rules
# while the user believes their ignores applied.
warnings.append({"path": ".wikiignore", "error": "ignore_unreadable"})
return []
return [line.strip() for line in text.splitlines() if line.strip() and not line.strip().startswith("#")]
def _ignored(rel: str, patterns: list[str]) -> bool:
return any(fnmatch.fnmatch(rel, pattern) or fnmatch.fnmatch(rel + "/", pattern) for pattern in patterns)
def _excluded(rel: str, patterns: list[str]) -> bool:
parts = rel.split("/")
return any(part in EXCLUDED_DIRS for part in parts) or _ignored(rel, patterns)
def _collect_sources(vault: str | Path) -> tuple[list[Path], list[dict[str, str]]]:
root = Path(vault).resolve()
warnings: list[dict[str, str]] = []
patterns = _ignore_patterns(root, warnings)
found: list[Path] = []
skipped: list[dict[str, str]] = warnings
# Enumerate without entering links, and report the links that were skipped. Every
# later read of a collected path is verified again, so a directory swapped into a
# link after enumeration is caught when its content is read.
for entry in boundary.walk(root, include_links=True):
rel = config.normalize_relpath(entry.path.relative_to(root).as_posix())
if entry.is_link:
skipped.append({"path": rel, "error": "skipped_symlink"})
continue
if not entry.name.endswith(".md"):
continue
if _excluded(rel, patterns):
continue
found.append(entry.path)
found.sort(key=lambda p: config.to_rel_posix(p, root))
skipped.sort(key=lambda item: item["path"])
return found, skipped
def walk_sources(vault: str | Path) -> Iterator[Path]:
boundary.bind(vault)
yield from _collect_sources(vault)[0]
def _item(path: Path, vault: Path, derived_topics: list[str] | None = None, *, cached_entry: dict[str, Any] | None = None) -> dict[str, Any]:
st = boundary.stat(path)
if st is None:
raise FileNotFoundError(2, "No such file or directory", str(path))
mtime_ns, size = st.st_mtime_ns, st.st_size
if cached_entry is not None:
cached_mtime_ns = cached_entry.get("mtime_ns")
cached_size = cached_entry.get("size")
if cached_mtime_ns is not None and cached_mtime_ns == mtime_ns and cached_size == size:
return {
"path": config.to_rel_posix(path, vault),
"size": size,
"mtime_ns": mtime_ns,
"derived_topics": derived_topics or [],
"signature_match": True,
}
sha256 = cache_mod.sha256_file(path)
return {
"path": config.to_rel_posix(path, vault),
"sha256": sha256,
"size": size,
"mtime_ns": mtime_ns,
"derived_topics": derived_topics or [],
}
def classify(vault: str | Path, cache_data: dict[str, Any], *, collect: Callable[[Path], tuple[list[Path], list[dict[str, str]]]] | None = None) -> dict[str, Any]:
boundary.bind(vault)
root = Path(vault).resolve()
sources = cache_data.get("sources", {})
result: dict[str, Any] = {"new": [], "modified": [], "unchanged": [], "deleted": [], "errors": [], "skipped_symlinks": []}
if collect is None:
source_paths, skipped = _collect_sources(root)
else:
source_paths, skipped = collect(root)
result["skipped_symlinks"] = skipped
result["errors"].extend(skipped)
current_paths: dict[str, Path] = {}
for path in source_paths:
rel = config.to_rel_posix(path, root)
if rel in current_paths:
result["fatal"] = {"error": "normalized_path_collision", "path": rel}
result["new"] = []
result["modified"] = []
result["unchanged"] = []
result["deleted"] = []
return result
current_paths[rel] = path
for rel, path in current_paths.items():
cached = sources.get(rel)
try:
item = _item(path, root, cached.get("derived_topics", []) if cached else [], cached_entry=cached)
except OSError:
result["errors"].append({"path": rel, "error": "unreadable"})
continue
if cached is None:
result["new"].append(item)
elif item.get("signature_match"):
result["unchanged"].append(item)
elif item["sha256"] != cached.get("sha256"):
result["modified"].append(item)
else:
result["unchanged"].append(item)
for rel, cached in sorted(sources.items()):
normalized = config.normalize_relpath(rel)
if normalized not in current_paths:
result["deleted"].append({"path": normalized, "derived_topics": cached.get("derived_topics", [])})
return result
def format_report(classified: dict[str, Any], vault: str | Path) -> dict[str, Any]:
if "fatal" in classified:
fatal_value = classified["fatal"]
if isinstance(fatal_value, dict):
return fatal_value
return {"error": str(fatal_value)}
return {
"version": 1,
"vault": str(Path(vault).resolve()),
"stats": {
"total_sources": len(classified.get("new", [])) + len(classified.get("modified", [])) + len(classified.get("unchanged", [])),
"new": len(classified.get("new", [])),
"modified": len(classified.get("modified", [])),
"unchanged": len(classified.get("unchanged", [])),
"deleted": len(classified.get("deleted", [])),
"skipped_symlinks": len(classified.get("skipped_symlinks", [])),
"errors": len(classified.get("errors", [])),
},
"new": classified.get("new", []),
"modified": classified.get("modified", []),
"deleted": classified.get("deleted", []),
"errors": classified.get("errors", []),
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |