Skip to main content
ClaudeWave
Skill14.8k repo starsupdated today

wiki-retrieve

**wiki-retrieve** is a chunk-level hybrid retrieval system for the Compound Vault that combines contextual prefixes, BM25 sparse search, and cosine reranking to replace the v1.6 page-level read order. Based on Anthropic's September 2024 Contextual Retrieval research, it reduces retrieval failure by 35–67% and is opt-in via setup script, with data privacy controls that allow fully on-machine operation using synthetic prefixes and local Ollama reranking.

Install in Claude Code
Copy
git clone --depth 1 https://github.com/AgriciDaniel/claude-obsidian /tmp/wiki-retrieve && cp -r /tmp/wiki-retrieve/skills/wiki-retrieve ~/.claude/skills/wiki-retrieve
Then start a new Claude Code session; the skill loads automatically.

SKILL.md

# Retrieve relevant passages

This extension derives search data from `wiki/` into `.vault-meta/`. It never
changes canonical notes. Always pass the selected vault explicitly.

Resolve the installed product root from this skill's own location, not from the
vault or current working directory:

```bash
PRODUCT_ROOT=/absolute/path/to/installed/claude-obsidian
PREFIX="$PRODUCT_ROOT/scripts/contextual-prefix.py"
BM25="$PRODUCT_ROOT/scripts/bm25-index.py"
RETRIEVE="$PRODUCT_ROOT/scripts/retrieve.py"
RERANK="$PRODUCT_ROOT/scripts/rerank.py"
test -f "$PREFIX" && test -f "$BM25" && test -f "$RETRIEVE" && test -f "$RERANK"
```

## Pipeline

1. `contextual-prefix.py` splits pages on paragraph boundaries and stores the
   raw chunk plus a short page-level prefix.
2. `bm25-index.py` builds a local, standard-library BM25 index over the
   contextualized text.
3. `retrieve.py` selects BM25 candidates, optionally reranks them, rejects
   invalid records, deduplicates by page, and returns paths and snippets.
4. The caller reads the returned pages and performs synthesis; retrieval output
   is not itself evidence.

## Provision locally

Preview first, then build synthetic prefixes without network egress:

```bash
python3 "$PREFIX" --vault "$VAULT" --all --no-llm --peek
python3 "$PREFIX" --vault "$VAULT" --all --no-llm
python3 "$BM25" --vault "$VAULT" build
python3 "$RETRIEVE" --vault "$VAULT" "wiki" --top 1 --no-rerank --explain
```

Chunk and index files are disposable runtime state. Incremental prefixing skips
records whose chunk and page hashes still match. A complete scan removes
surplus records for deleted pages, and the prefixer invalidates the BM25 index
before changing its chunk set so a mixed stale index is not served.
Prefix and BM25 build operations share the vault-wide mutation lock with every
other writer; a busy vault fails closed instead of publishing a partial index.

## Contextual-prefix privacy

Synthetic prefixes use only local frontmatter and page text. The Anthropic API
and `claude` subprocess tiers can send page bodies off-machine and therefore
require the user's explicit consent plus `--allow-egress`. Never infer consent
from an API key or installed binary. Preview the scope first and state which
provider will receive what data.

Remote Ollama endpoints also require explicit approval and
`--allow-remote-ollama`; the default reranker accepts localhost only.

## Query

For a strictly read-only lookup, use the prebuilt BM25 index:

```bash
python3 "$RETRIEVE" --vault "$VAULT" "$QUERY" --top 5 --no-rerank --explain
```

For an explicitly requested rerank, omit `--no-rerank`. The default is Ollama's
multilingual `nomic-embed-text-v2-moe` model (approximately 958 MB); the product
never pulls it automatically. To use an already-installed, smaller,
English-oriented v1.5 model, pass `--model nomic-embed-text` explicitly.
Nomic models use `search_query:` for the query and `search_document:` for
candidate text. Nomic v2 has a 512-token input context and Ollama truncates
longer embedding inputs by default; BM25 still scores the complete chunk.
Embeddings are cached by exact model, input scheme, and hash of the exact
prefixed input. A missing local Ollama service, missing selected
model, unusable vector, or any candidate embedding failure falls back for the
complete result set to the original BM25 order; it never mixes cosine and BM25
score scales.

Query input is bounded at 8,000 normalized characters and result counts must be
between 1 and 1,000. Oversized queries and invalid limits fail with an
actionable usage error instead of looking like an empty successful search.
An untagged model request matches only the installed untagged name or its
`:latest` alias; select any other tag explicitly.

Use direct diagnostics when needed:

```bash
python3 "$BM25" --vault "$VAULT" stats
python3 "$BM25" --vault "$VAULT" query "$QUERY" --top 10
python3 "$RERANK" --vault "$VAULT" "$QUERY" --peek
python3 "$RERANK" --vault "$VAULT" "$QUERY" --model nomic-embed-text --peek
```

## Integrity rules

- Accept only relative chunk and page paths whose resolved targets remain under
  `$VAULT/.vault-meta/chunks/` and `$VAULT/wiki/` respectively.
- Reject hashless legacy chunk records and require chunk-body, page, and index
  hashes to match before a cached record can be built or served.
- Reject absolute paths, symlink escapes, missing pages, mismatched chunk IDs,
  changed page hashes, and stale index/chunk hash pairs.
- Rerank the full candidate set, then deduplicate by page, then apply `--top`.
- An empty index is an honest no-result state. A missing or corrupt index makes
  `retrieve.py` exit 10 with a stable rebuild command; callers fall back to the
  standard vault query/text-search path and do not fabricate matches.
- Do not cite benchmark percentages unless a reproducible vault-specific
  benchmark produced them.

## Checkpoint

Observe cache readiness and privacy boundaries, think about whether lexical or
semantic ranking is needed, verify returned paths and source freshness, and
grow by measuring retrieval misses against a maintained local query set.
verifierSubagent

>

wiki-ingestSkill

Ingest supplied source material into an Obsidian vault with provenance and claim tracking: pasted text, files staged in the selected vault's inbox or .raw archive, or explicitly approved URLs. Use for a single source or bounded batch, not for saving an assistant answer. Triggers: ingest, ingest this file, ingest this URL, process this source, read and file this source, batch ingest, ingest these sources.

wiki-lintSkill

Run a deterministic, read-only health check on an Obsidian wiki. Use for lint, vault health check, audit wiki health, find orphans, find dead links, frontmatter audit, provenance audit, or wiki audit. Reports graph, link, frontmatter, provenance-ledger, empty-section, and stale-index findings; it does not reason broadly or repair files.

autoresearchSkill

Run a bounded, source-grounded research loop, draft a cited dossier, and optionally propose a separately reviewed canonical vault merge. Use when the user wants autonomous or deep research that may access the public web. Triggers: /autoresearch, autoresearch, research this topic, deep dive into, investigate, find everything about, research and file, go research, build a wiki on.

canvasSkill

Create, inspect, and update Obsidian JSON Canvas boards with text, file, link, group, and edge nodes. Use for canvas status, canvas lists, visual maps, zones, spatial layouts, adding vault notes or media to a .canvas file, and requests such as create canvas, add to canvas, or put this on the canvas.

saveSkill

Save a user-selected answer, decision, insight, or session summary into an Obsidian vault as one reviewed transaction. Use only when the user explicitly asks to preserve specific conversation content, not when they supply a file or URL to ingest. Triggers: /save, save this, save that answer, file this conversation, save this analysis, keep this insight, preserve this chat result.

wikiSkill

Initialize, adopt, and route work for a separate Obsidian knowledge vault through the portable claude-obsidian core. Use for vault setup, scaffolding, workspace selection, cross-project configuration, or choosing the correct wiki sub-skill. Triggers: /wiki, set up wiki, scaffold vault, create knowledge base, adopt this vault, Obsidian vault, second brain setup, persistent wiki.

defuddleSkill

Plan and, with explicit network consent, use an optional external Defuddle cleaner to extract article-like HTTPS pages as Markdown. Use for defuddle, clean this URL, strip page clutter, readable Markdown from a web page, or preparing a web source for later wiki ingestion.