#!/usr/bin/env python3
"""Add provenance frontmatter (author/adapted-from) + missing versions to all skills/.

Idempotent: only inserts lines that are absent. Preserves all existing content.
Usage: python scripts/add-provenance.py   (run from repo root)
"""
import os
import re
import sys

ROOT = os.path.join(os.path.dirname(os.path.abspath(__file__)), "..")
SKILLS = os.path.join(ROOT, "skills")

# skill -> provenance line(s) to guarantee present in frontmatter
PROVENANCE = {
    # ours — authored in this workshop/ecosystem
    "active-platform": ["author: ours", "version: 0.1.0"],
    "browser-cdp": ["author: ours"],
    "delegate-task": ["author: ours"],
    "inter-session-peer-discipline": ["author: ours"],
    "meta-host-routing": ["author: ours"],
    "private-dev-public-publish": ["author: ours"],
    "project-bootstrap": ["author: ours"],
    "project-discipline": ["author: ours"],
    "pulling-before-work": ["author: ours"],
    "ralph-loop-execution": ["author: ours"],
    "recommend-dont-menu": ["author: ours"],
    "session-handoff": ["author: ours"],
    "session-inbox-monitor": ["author: ours"],
    "setup-agents-task-runner": ["author: ours"],
    "setup-context7": ["author: ours"],
    "setup-interns": ["author: ours"],
    "setup-projects-meta": ["author: ours"],
    "setup-tasks": ["author: ours"],
    "setup-wiki": ["author: ours"],
    "task-format": ["author: ours"],
    "task-loop": ["author: ours"],
    "tdd-criteria": ["author: ours"],
    "update-skills": ["author: ours"],
    "using-context7": ["author: ours"],
    "using-interns": ["author: ours"],
    "using-markitdown": ["author: ours"],
    "using-projects-meta": ["author: ours"],
    "using-system-snapshot": ["author: ours"],
    "using-tasks": ["author: ours"],
    "using-vds-ops": ["author: ours"],
    "using-wiki": ["author: ours"],
    "using-wiki-graph": ["author: ours"],
    # adapted — vendored copies, sovereignly maintained; pins to backfill (see README "Adapted from")
    "caveman": ["adapted-from: JuliusBrussee/caveman (MIT) — vendored copy, upstream pin TBD", "version: 0.1.0"],
    "caveman-commit": ["adapted-from: JuliusBrussee/caveman (MIT) — vendored copy, upstream pin TBD", "version: 0.1.0"],
    "caveman-compress": ["adapted-from: JuliusBrussee/caveman (MIT) — vendored copy, upstream pin TBD"],
    "caveman-help": ["adapted-from: JuliusBrussee/caveman (MIT) — vendored copy, upstream pin TBD", "version: 0.1.0"],
    "caveman-review": ["adapted-from: JuliusBrussee/caveman (MIT) — vendored copy, upstream pin TBD", "version: 0.1.0"],
    "find-skills": ["adapted-from: vercel-labs/skills (MIT) — vendored copy, upstream pin TBD", "version: 0.1.0"],
}

FRONTMATTER_RE = re.compile(r"^(---\s*\n)(.*?)(\n---\s*\n)", re.DOTALL)

changed = []
errors = []
for name, lines in sorted(PROVENANCE.items()):
    path = os.path.join(SKILLS, name, "SKILL.md")
    if not os.path.exists(path):
        errors.append(f"MISSING: {path}")
        continue
    with open(path, encoding="utf-8") as f:
        text = f.read()
    m = FRONTMATTER_RE.match(text)
    if not m:
        errors.append(f"{name}: no frontmatter")
        continue
    fm = m.group(2)
    new_lines = []
    inserted = False
    for line in fm.splitlines():
        # insert provenance after the name: line (keep name-first convention)
        if not inserted and line.startswith("name:"):
            new_lines.append(line)
            for prov in lines:
                key = prov.split(":", 1)[0]
                # skip if this key already present anywhere in frontmatter
                if not any(existing.startswith(key + ":") for existing in fm.splitlines()):
                    new_lines.append(prov)
                    inserted = True
            continue
        # drop old lines that we're replacing (e.g. duplicate version)
        new_lines.append(line)
    # dedupe: remove lines for keys that now appear twice
    seen = {}
    final = []
    for line in new_lines:
        key = line.split(":", 1)[0]
        if key in ("name",):
            final.append(line)
            continue
        if key in seen and seen[key] >= 1:
            continue  # keep first occurrence
        seen[key] = seen.get(key, 0) + 1
        final.append(line)
    new_fm = "\n".join(final)
    if new_fm != fm:
        new_text = text[: m.start(2)] + new_fm + text[m.end(2):]
        with open(path, "w", encoding="utf-8", newline="\n") as f:
            f.write(new_text)
        changed.append(name)
    else:
        print(f"  = {name}: no change")

print(f"\nChanged ({len(changed)}): {', '.join(changed)}")
if errors:
    print(f"\nERRORS ({len(errors)}):")
    for e in errors:
        print(" ", e)
    sys.exit(1)