wiki-ingest: Source Ingestion

Read the source. Write the wiki. Cross-reference everything. A single source typically touches 8-15 wiki pages.

Syntax standard: Write all Obsidian Markdown using proper Obsidian Flavored Markdown. Wikilinks as [[Note Name]], callouts as > [!type] Title, embeds as ![[file]], properties as YAML frontmatter. If the kepano/obsidian-skills plugin is installed, prefer its canonical obsidian-markdown skill for Obsidian syntax reference. Otherwise, follow the guidance in this skill.

Transport (v1.7+)

Before mutating any vault file, consult .vault-meta/transport.json (auto-created by bash scripts/detect-transport.sh). Use the preferred transport per the fallback chain:

• cli — obsidian-cli write "$VAULT" "$NOTE" < content.md (or append, property:set); see skills/wiki-cli/SKILL.md
• mcp-obsidian / mcpvault — mcp__obsidian-vault__write_note and friends; see skills/wiki/references/mcp-setup.md
• filesystem — Claude's Write/Edit tools with absolute vault-rooted paths (final floor; always works)

Full decision tree: wiki/references/transport-fallback.md.

Mode awareness (v1.8+)

Before creating any new wiki page, consult the vault's methodology mode via python3 scripts/wiki-mode.py route <type> "<name>". The router returns the vault-relative path where the page should be filed.

SRC_PATH=$(python3 scripts/wiki-mode.py route source "Karpathy 2025 LLM Wiki essay")
# generic:      wiki/sources/Karpathy-2025-LLM-Wiki-essay.md
# lyt:          wiki/notes/Karpathy-2025-LLM-Wiki-essay.md  (also update relevant MOC)
# para:         wiki/resources/incoming/Karpathy-2025-LLM-Wiki-essay.md
# zettelkasten: wiki/20260517123456-Karpathy-2025-LLM-Wiki-essay.md

ENT_PATH=$(python3 scripts/wiki-mode.py route entity "Andrej Karpathy")
CON_PATH=$(python3 scripts/wiki-mode.py route concept "Compounding Vault Pattern")

If .vault-meta/mode.json is absent, the router returns mode=generic paths (identical to v1.7 behavior). No special-casing needed in this skill.

Mode-specific follow-up:

• LYT: after filing the atomic note, update the relevant MOC (wiki/mocs/<topic>-moc.md) to link the new note. If no MOC exists for the topic, create one using skills/wiki-mode/templates/lyt/moc-template.md.
• Zettelkasten: filename already includes the timestamp ID. Populate the id: frontmatter field to match.
• PARA: new ingests land in wiki/resources/incoming/ by default. Do NOT auto-guess the topic; leave in incoming/ for user review.

Concurrency (v1.7+)

Multi-writer is safe in v1.7. The latent corruption bug from v1.6 — where two parallel sub-agents writing to the same page could silently trample each other — is closed by per-file advisory locking. Every wiki page write MUST be preceded by wiki-lock acquire <path>.

# Acquire — blocks (returns 75 EX_TEMPFAIL) if another writer holds the lock
if bash scripts/wiki-lock.sh acquire wiki/concepts/Foo.md; then
  # ... do the write via the §Transport-selected method ...
  bash scripts/wiki-lock.sh release wiki/concepts/Foo.md
else
  # rc=75: another writer is in flight. Retry once after 2s; if still held,
  # log to wiki/log.md and skip this page rather than overwrite.
  sleep 2
  bash scripts/wiki-lock.sh acquire wiki/concepts/Foo.md && {
    # write …
    bash scripts/wiki-lock.sh release wiki/concepts/Foo.md
  } || echo "skipped wiki/concepts/Foo.md (locked); logged to wiki/log.md"
fi

Properties:

• Per-file granularity. Locks key on sha1(<vault-relative-path>); concurrent writes to DIFFERENT pages run in parallel.
• Age-based staleness. Default STALE_AFTER_SEC=60. A crashed holder unblocks in ≤60 seconds without manual intervention. See scripts/wiki-lock.sh header for the full semantics.
• Cross-process release. Release is rm -f (no PID match required). Skill authors are trusted to release locks they acquire; cross-skill release is allowed by design (a janitor running wiki-lock clear-stale --max-age 0 is the canonical recovery path).
• The PostToolUse hook now defers git add if any locks are currently held, so the auto-commit doesn't fire mid-ingest and produce torn commits. See hooks/hooks.json.

wiki-lock is unconditional in v1.7+ — there is no feature gate, no fallback. Skills that don't acquire locks are racing against any other writer. The script is in core, not opt-in.

Sub-agent rule from v1.6 — "Sub-agents MUST NOT call scripts/allocate-address.sh" — is preserved (orchestrator still backfills addresses to keep the counter monotonic). The NEW rule is: sub-agents MAY now write pages, but MUST acquire locks first. See agents/wiki-ingest.md.

Delta Tracking

Before ingesting any file, check .raw/.manifest.json to avoid re-processing unchanged sources.

# Check if manifest exists
[ -f .raw/.manifest.json ] && echo "exists" || echo "no manifest yet"

Manifest format (create if missing):

{
  "sources": {
    ".raw/articles/article-slug-2026-04-08.md": {
      "hash": "abc123",
      "ingested_at": "2026-04-08",
      "pages_created": ["wiki/sources/article-slug.md", "wiki/entities/Person.md"],
      "pages_updated": ["wiki/index.md"]
    }
  }
}

Before ingesting a file:

1. Compute a hash: md5sum [file] | cut -d' ' -f1 (or sha256sum on Linux).
2. Check if the path exists in .manifest.json with the same hash.
3. If hash matches, skip. Report: "Already ingested (unchanged). Use force to re-ingest."
4. If missing or hash differs, proceed with ingest.

After ingesting a file:

1. Record {hash, ingested_at, pages_created, pages_updated} in .manifest.json.
2. Write the updated manifest back.

Skip delta checking if the user says "force ingest" or "re-ingest".

URL Ingestion

Trigger: user passes a URL starting with https://.

Steps:

1. Fetch the page using WebFetch.
2. Clean (optional): if defuddle is available (which defuddle 2>/dev/null), run defuddle [url] to strip ads, nav, and clutter. Typically saves 40-60% tokens. Fall back to raw WebFetch output if not installed.
3. Derive slug from the URL path (last segment, lowercased, spaces→hyphens, strip query strings).
4. Save to .raw/articles/[slug]-[YYYY-MM-DD].md with a frontmatter header:

   ---
   source_url: [url]
   fetched: [YYYY-MM-DD]
   ---
   

5. Proceed with Single Source Ingest starting at step 2 (file is now in .raw/).

Image / Vision Ingestion

Trigger: user passes an image file path (.png, .jpg, .jpeg, .gif, .webp, .svg, .avif).

Steps:

1. Read the image file using the Read tool. Claude can process images natively.
2. Describe the image contents: extract all text (OCR), identify key concepts, entities, diagrams, and data visible in the image.
3. Save the description to .raw/images/[slug]-[YYYY-MM-DD].md:

   ---
   source_type: image
   original_file: [original path]
   fetched: YYYY-MM-DD
   ---
   # Image: [slug]
   
   [Full description of image contents, transcribed text, entities visible, etc.]
   

4. Copy the image to _attachments/images/[slug].[ext] if it's not already in the vault.
5. Proceed with Single Source Ingest on the saved description file.

Use cases: whiteboard photos, screenshots, diagrams, infographics, document scans.

Single Source Ingest

Trigger: user drops a file into .raw/ or pastes content.

Steps:

1. Read the source completely. Do not skim.
2. Discuss key takeaways with the user. Ask: "What should I emphasize? How granular?" Skip this if the user says "just ingest it."
3. Create source summary in wiki/sources/. Use the source frontmatter schema from references/frontmatter.md. Assign an address per the Address Assignment section below.
4. Create or update e