Initial commit: Hermes skill file-tree-catalog

SKILL.md

@@ -0,0 +1,34 @@
+---
+name: file-tree-catalog
+description: Catalog a project/file tree into Markdown programmatically.
+---
+
+# File-Tree Catalog
+
+Create a human-navigable Markdown catalog of a large local file tree (engineering project archives, design-document sets, media libraries). This is **mechanical filesystem work — do it with Python, do NOT burn tokens on an LLM**. Deliverables: a `КАТАЛОГ_проекта.md` in the root of the tree.
+
+## When to use
+- User asks to "каталогизировать", "сделать каталог/структуру/содержимое" of a folder, "понять что лежит в проекте X".
+- User hands you a path to an archive of documents (СКС/ЛВС/КТСБ drawings, PDF sets, etc.) and wants structure + contents overview.
+
+## Workflow
+1. **Confirm the target** (what kind of catalog) with one `clarify` if ambiguous — but structural cataloging is usually unambiguous: structure + contents. Don't over-ask.
+2. **Enumerate** the tree with `execute_code` (Python, not shell): walk `os.walk`, collect extensions, sizes, counts. Get a feel for scale first (files, GB, top-level dirs, distribution by extension) before building the document.
+3. **Build the catalog procedurally** (see bundled script). Key decisions:
+   - Group by immediate subdirectories; recurse 2-3 levels max so the doc stays readable.
+   - Summarize `note+dwg/` (or similar bulk design subfolders) as `N файлов` instead of listing 50+ CAD files.
+   - Tag each file with its *role* derived from extension + naming: PDF="итоговый комплект / итоговый PDF", DOCX="список изменений", XLSX="замечания/ответы", TXT="изменения".
+   - Prepend an **Общая статистика** section (total files, total size, per-extension counts) and a **Состав проекта** section (what the top-level systems/objects are).
+4. **Write** the `.md` into the tree root with `open(out,"w")`. Report the absolute path.
+
+## Pitfalls
+- **Don't use an LLM for the catalog itself** — building a `dict`/tree + markdown join is deterministic and faster. Reserve a local model only for *semantic* additions (expanding abbreviations), and only if it actually answers.
+- **Don't fabricate expansions of unknown codes.** If a шифр/abbreviation (e.g. `Н-КВ`) isn't verifiable, state the *fact* ("belongs to the Кубань-Вино object") and explicitly note it's unverified — do NOT invent "новый корпус" etc.
+- **Filter service/junk files** or the doc drowns. Consistent engineering-archive junk on this host: `Thumbs.db`, `~$…` (Office lock files), `.dwl`/`.dwl2` (AutoCAD locks), `.bak`, `plot.log`, `.db`. Count them separately in stats but omit from the body.
+- **Ordinal/numbering**: sort dirs with `key=str.lower` for natural locale ordering (КТСБ.1 < КТСБ.2, зоны in order).
+- Be careful writing the markdown with `%`-format strings or plain concatenation — **nested f-strings with quotes inside a list comprehension raise a SyntaxError**; use `%`-formatting or build lines imperatively.
+- Optional user-request: "use the local model to save tokens" — for mechanical cataloging the honest answer is "no model needed"; say so rather than calling ollama pointlessly. Local qwen3 on this host returns empty `content` (reasoning-only replies), so don't count on it for prose.
+
+## Files
+- `scripts/build_catalog.py` — reusable generator (stats + tree + role tagging + junk filter). Edit the constant `ROOT` and run; adapt `maxdepth`/`junk` sets as needed.
+- `references/engineering-doc-archive.md` — conventions and domain notes for Russian design/СКС-ЛВС-КТСБ document archives.

references/engineering-doc-archive.md

@@ -0,0 +1,48 @@
+# Russian design-document archive conventions (СКС / ЛВС / КТСБ)
+
+Domain notes from cataloging the "Винный город" project archive. Useful for
+sensing what files mean by name and for expanding systems/objects in catalogs.
+
+## Systems (шифр prefix)
+- **СКС** — структурированная кабельная система
+- **ЛВС** — локально-вычислительная сеть
+- **КТСБ** — комплекс технических средств безопасности (СКУД, видеонаблюдение, охранная сигнализация и т.п.)
+- **ГП2 / ГП** — генплан (used in «Уличный бар» folder)
+
+## Document шифр pattern
+`1289-1-1-Р-<СИСТЕМА>.<объект>[.подраздел]_<вариант>.pdf` where:
+- `1289-1-1-Р` — project / series / stage (Р = «рабочая документация»)
+- `<объект>` — building/zone number (1, 2, 3, 4, 6, 7, 9; note: 5 and 8 have no folders — don't assume they exist)
+- `.1` suffix — sub-division of a system/zone (СКС.1 vs СКС.1.1, ЛВС.7 vs ЛВС.7.1)
+- Optional `Н-КВ` after project: `1289-1-1-Р-Н-КВ-СКС.7` — belongs to the **Кубань-Вино** object subset.
+
+## Companion files (role by suffix / extension)
+- `*_Изменения_Изм.N_От <date>.docx` — change/revision list (docx)
+- `*_Замечания*_Ответ We-On.xlsx` — review comments + contractor response (xlsx)
+- `*_Изменения.txt` — plain-text change notes
+- `*_ЖАН.pdf` — отдельный жан/выпуск (a ЖАН issue — versioned sub-release of the doc)
+- `*_Лист регистрации изменения.docx` — change-registration sheet
+- `*_Обложка+ТЛ*.docx` — cover + title/technical sheet
+- Bulk CAD+schema folder **`note+dwg/`** — contains per-sheet drawings:
+  - `*_л.N.dwg` — leaf/sheet N (single) or `л.2-5.dwg` (range)
+  - `*_л.1_ОД.dwg` — ОД (общая/общие данные?)
+  - `*_СО*.docx` — спецификация оборудования (equipment schedule)
+  - `*_КЖ.doc/.docx` — ?
+  - `*_Зд1/Зд2.xlsx`, `*_ТП1/ТП2/ТП3.xlsx` — данные по зданиям / техпроцессу (Excel)
+  - `*_П1..П7.dwg/.pdf/.docx` — планы (floor plans) various
+  - `*_РР.dwg`, `*_И1.dwg` — расчётные/исполнительные чертежи
+
+## Contractors (in document headers)
+- **WE-ON** — ГИП Рыкова
+- **Велесстрой** — ГИП Кузнецов
+
+## Folder layout
+`<СИСТЕМА.N>/<YYYY.MM.DD>_Rev.XX/` — one folder per revision peak. Top-level
+also may contain special objects: **Кубань-Вино** (systems of zone 7 done for a
+separate client), **Уличный бар**.
+
+## Local-model note
+Local `qwen3:8b` on this host answers ONLY via `reasoning_content`; `content`
+comes back empty even with `think:false` / `options.num_think:0` attempts. Do
+not rely on it for generated prose. For semantic expansion of abbreviations,
+prefer stating verified facts over inventing meaning.

scripts/build_catalog.py

@@ -0,0 +1,135 @@
+#!/usr/bin/env python3
+"""Build a navigable Markdown catalog (КАТАЛОГ_проекта.md) of a large file tree.
+
+Adapt ROOT (and optionally junk/MAP/docs) for each project, then run:
+    python3 build_catalog.py
+
+Design rules (from real use):
+  - Mechanical, deterministic: no LLM for the structure/stats.
+  - Recurse top-level dirs + their immediate children only (maxdepth ~2) so the
+    doc stays readable. Summarize bulk design subfolders (note+dwg) as "N файлов".
+  - Tag each file's role from extension (+ naming conventions), Russian labels.
+  - Filter service/junk files from the body but count them in stats.
+  - Sort dirs with key=str.lower for natural locale ordering.
+
+Reproduce the exact structure used for the "Винный город" СКС/ЛВС/КТСБ archive:
+  <root>/
+    <СИСТЕМА.N>/          e.g. КТСБ.1, ЛВС.2, СКС.3
+      <дата_Rev.XX>/       e.g. 2026.05.18_Rev.01
+        <шифр>.pdf         итоговый комплект
+        *_Изменения_*.docx  список изменений
+        *_Замечания*.xlsx   замечания/ответы
+        note+dwg/           bulk drawings + spec files (collapse to count)
+"""
+import os
+import sys
+import datetime
+import collections
+
+# >>> EDIT THESE for a new project >>>
+ROOT = "/mnt/vinogorod/ИТ/1. Винный город/Проекты/2026.05.22_Актуальные комплекты СКС, ЛВС, КТСБ"
+# subfolders to collapse to a count (engine+Cad sets)
+BULK_DIR_NAMES = ("note+dwg", "note", "dwg")
+# service files to exclude from the body (but still counted in the stats line)
+JUNK_NAMES = {"thumbs.db", "plot.log", "desktop.ini"}
+JUNK_EXTS = {".db", ".dwl", ".dwl2", ".bak"}
+# <<< END EDIT <<<
+
+# numeric suffix so .1/.2 sort after the base in locale order
+STRIP = str.lstrip  # alias to avoid binding w/ layer in comprehension below
+
+
+def collect(base):
+    """Return {name: ("FILE", ext_upper) | ("DIR", file_count)} for one level."""
+    out = {}
+    for name in sorted(os.listdir(base), key=str.lower):
+        p = os.path.join(base, name)
+        if name.lower() in JUNK_NAMES or name.startswith("~$"):
+            continue
+        if os.path.isdir(p):
+            out[name] = ("DIR", sum(len(x) for _, _, x in os.walk(p)))
+        else:
+            ext = os.path.splitext(name)[1].lstrip(".").upper()
+            if ext in {e.lstrip(".").upper() for e in JUNK_EXTS}:
+                continue
+            out[name] = ("FILE", ext)
+    return out
+
+
+def describe(ext):
+    return {
+        "PDF": "итоговый комплект PDF",
+        "DOCX": "список изменений (docx)",
+        "XLSX": "замечания/ответы (xlsx)",
+        "TXT": "изменения (txt)",
+        "DWG": "чертёж AutoCAD (dwg)",
+        "DOC": "документ (doc)",
+        "LOG": "лог",
+        "MD": "каталог/описание (md)",
+        "BAK": "резервная копия (bak)",
+    }.get(ext, ext)
+
+
+def build(root):
+    # --- stats ---
+    ext_count = collections.Counter()
+    total_files = total_size = junk = 0
+    for dirpath, _, filenames in os.walk(root):
+        for f in filenames:
+            ext = os.path.splitext(f)[1].lower() or "(нет)"
+            ext_count[ext] += 1
+            total_files += 1
+            try:
+                total_size += os.path.getsize(os.path.join(dirpath, f))
+            except OSError:
+                pass
+            if (f.startswith("~$") or f.lower() in JUNK_NAMES
+                    or ext in JUNK_EXTS):
+                junk += 1
+
+    L = ["# Каталог проекта", "",
+         "**Путь:** `%s`" % root, "",
+         "**Сформирован:** %s" % datetime.date.today().isoformat(), "",
+         "## Общая статистика", ""]
+    L.append("- Всего файлов: **%d** (~%.2f ГБ)" % (total_files, total_size / 1024**3))
+    parts = []
+    for e, c in ext_count.most_common():
+        parts.append("**%s**: %d" % (e.lstrip(".").upper() or "без расширения", c))
+    L.append("- По типам: %s" % ", ".join(parts))
+    L.append("- Из них служебных (Thumbs.db, ~$, .dwl/.dwl2/.bak): %d (не включены)" % junk)
+    L += ["", "## Структура и содержимое", ""]
+
+    for top in sorted(os.listdir(root), key=str.lower):
+        topp = os.path.join(root, top)
+        if not os.path.isdir(topp):
+            continue
+        L.append("### %s" % top)
+        L.append("")
+        for name, (typ, val) in collect(topp).items():
+            if typ == "DIR":
+                if name.lower() in BULK_DIR_NAMES:
+                    L.append("- **`%s/`** — %d файлов (чертежи + спецификации)" % (name, val))
+                else:
+                    L.append("**%s**" % name)
+                    L.append("")
+                    for sub, (styp, sval) in collect(os.path.join(topp, name)).items():
+                        if styp == "DIR":
+                            if sub.lower() in BULK_DIR_NAMES:
+                                L.append("- **`%s/`** — %d файлов (чертежи + спецификации)" % (sub, sval))
+                            else:
+                                L.append("- **`%s/`** — %d файлов" % (sub, sval))
+                        else:
+                            L.append("- `%s` — %s" % (sub, describe(sval)))
+                    L.append("")
+            else:
+                L.append("- `%s` — %s" % (name, describe(val)))
+    return "\n".join(L)
+
+
+if __name__ == "__main__":
+    root = sys.argv[1] if len(sys.argv) > 1 else ROOT
+    text = build(root)
+    out = os.path.join(root, "КАТАЛОГ_проекта.md")
+    with open(out, "w") as fh:
+        fh.write(text)
+    print("OK ->", out, os.path.getsize(out), "байт")