wiki-retrieve — Claude Code skill for vault search
wiki-retrieve builds a local BM25 index from wiki/ notes and finds relevant passages. Use it when an agent needs matching pages to read before answering a question.
MIT 348 installs ★ 16,676 updated 2026-10-22
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a claude-codeWhat wiki-retrieve does
wiki-retrieve splits pages into chunks, adds a short page-level context prefix, then indexes the resulting text with BM25. Search returns paths and snippets; the agent reads those pages and produces its own synthesis. Index data and derived files live in .vault-meta, while canonical notes remain unchanged.
Optional reranking uses a multilingual Nomic model through Ollama. If the model or service is unavailable, results fall back to their original BM25 order. Remote Ollama endpoints and the Anthropic API require explicit consent and the relevant flags before page text can leave the machine. Run the scripts with Python 3 and pass the target vault using --vault.
What it helps with
- find relevant passages in
wiki/notes - build a local BM25 index for a vault
- rerank search candidates with a Nomic model
- inspect index statistics and BM25 results
Example prompts for the agent
- Search
wiki/for notes about retrieval setup and return five passages without reranking. - Build the index for this vault, then explain the results for the query
wiki. - Rerank the matches with
nomic-embed-textif that model is already installed in Ollama.
How to install wiki-retrieve
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a claude-code2Manually: put the files in place
~/.claude/skills/wiki-retrieve/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/wiki-retrieve/
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a codex2Manually: put the files in place
.agents/skills/wiki-retrieve/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a cursor2Manually: put the files in place
.cursor/skills/wiki-retrieve/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install AgriciDaniel/claude-obsidian skills/wiki-retrieve2Manually: put the files in place
.github/skills/wiki-retrieve/SKILL.md1Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a gemini-cli2Platform CLI
gemini skills install https://github.com/AgriciDaniel/claude-obsidian --path skills/wiki-retrieveManually: put the files in place
.devin/skills/wiki-retrieve/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about wiki-retrieve
How do I install wiki-retrieve in Claude Code?
Run npx skills add AgriciDaniel/claude-obsidian --skill wiki-retrieve -a claude-code. The skill file is available at ~/.claude/skills/wiki-retrieve/SKILL.md.
What does wiki-retrieve do?
It builds a local BM25 index from notes in wiki/ and returns matching passages with their page paths. The agent reads those pages and creates the answer separately.
Does search require an embedding model?
No. The base search uses BM25, and Ollama-based reranking is optional. If the selected model or service cannot be used, results retain their BM25 order.
Source text
SKILL.md · MIT
--- name: wiki-retrieve description: "Build and query a vault-local contextual BM25 retrieval index with optional multilingual Nomic cosine reranking; use for retrieve, hybrid retrieval, BM25, rerank, contextual retrieval, chunk search, vault search, semantic search, find relevant passages, or retrieval diagnostics. Derived caches stay under .vault-meta, remote egress requires explicit consent, and unavailable reranking falls back deterministically." --- # 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 hashes, page hashes and provider/model profile still match. Legacy records without that profile regenerate on an explicit build. 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 explicit Anthropic API profile can send page bodies off-machine and requires the user's consent plus `--allow-egress` and a configured API key. There is no Claude CLI generation tier or ambient subprocess fallback. 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. Local model prefixes are a separate explicit option. Use an already installed exact `name:tag`; no server is started and no model is installed or pulled: ```bash python3 "$PREFIX" --vault "$VAULT" --all --local-ollama --ollama-model installed:tag --peek python3 "$PREFIX" --vault "$VAULT" --all --local-ollama --ollama-model installed:tag ``` The prefix adapter verifies cloud-disabled server status and local model identity before sending page text. Unsupported or unavailable preflight falls back to synthetic prefixes for new or changed chunks only; prefixes from the same endpoint and model stay in place, so a transient outage leaves an unchanged vault's chunks and BM25 index untouched. This requirement can exclude older Ollama servers. Reranking and tiling use the same local-only proof before query, page, chunk, or embedding cache work. `--allow-remote-ollama` approves the endpoint, not cloud delegation by the server. See [prefix providers](../../docs/prefix-providers.md) for the exact policy, cloud model selection and verification limits. The reranker has its own model and endpoint policy; selecting local prefixes does not enable reranking. ## Query For a strictly read-only lookup, use the prebuilt BM25 index: ```bash python3 "$RETRIEVE" --vault "$VAULT" "$QUERY" --top 5 --no-rerank --explain ``` For an explicitly requested rerank, omit `--no-rerank`. The default is Ollama's multilingual `nomic-embed-text-v2-moe` model (approximately 958 MB); the product never pulls it automatically. To use an already-installed, smaller, English-oriented v1.5 model, pass `--model nomic-embed-text` explicitly. Nomic models use `search_query:` for the query and `search_document:` for candidate text. Nomic v2 has a 512-token input context and Ollama truncates longer embedding inputs by default; BM25 still scores the complete chunk. Embeddings are cached by exact model, verified digest, input scheme, and hash of the exact prefixed input. A missing local Ollama service, malformed HTTP reply, missing selected model, unusable vector, or any candidate embedding failure falls back for the complete result set to the original BM25 order; it never mixes cosine and BM25 score scales. Query input is bounded at 8,000 normalized characters and result counts must be between 1 and 1,000. Oversized queries and invalid limits fail with an actionable usage error instead of looking like an empty successful search. An untagged model request matches only the installed untagged name or its `:latest` alias; select any other tag explicitly. Use direct diagnostics when needed: ```bash python3 "$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. Measure a prebuilt index against explicit labels with `bench --queries FILE --vault VAULT`. See [offline retrieval evaluation](../../docs/retrieval-evaluation.md) for recall, MRR, missing labels and negative cases. This is a read-only measurement and never silently rebuilds an index or calls a model. ## 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.
© AgriciDaniel · MIT · AgriciDaniel/claude-obsidian