Skip to main content
Glama

Hybrid search (BM25 + TF-IDF + embeddings, RRF-fused)

obsidian_search
Read-onlyIdempotent

Retrieve relevant vault content by fusing full-text, semantic, and graph signals into one ranked list, with per-result explanation and granularity control.

Instructions

The default search tool for v2.0. Auto-detects every available retrieval signal — BM25 via FTS5 (if --persistent-index), TF-IDF cosine (always), and ML embeddings (if enquire-mcp build-embeddings ran) — and fuses them with Reciprocal Rank Fusion (Cormack et al, 2009) for higher recall and better paraphrase / synonym matching than any single ranker. Equal weights, k=60. Gracefully degrades: with only TF-IDF available it produces TF-IDF-style ranking; with BM25+TF-IDF it does keyword-augmented retrieval; with all 3 it matches Smart Connections-quality retrieval — free / offline / open-source. Returns per-signal observability (per_signal: { bm25, tfidf, embeddings }) so you can see WHY each hit ranked. v2.8.0: when --include-pdfs was passed to serve (or enquire-mcp index --include-pdfs ran), PDF chunks are blended into results — each hit carries a kind: "md" | "pdf" flag and PDF chunks include [page: N] markers in snippets so agents can cite the right page. Use this instead of the individual _search_text / _full_text_search / _semantic_search / _embeddings_search tools unless you specifically need single-ranker output for diagnostics. v3.10 (forgetting-aware): every hit also carries age_days (whole days since the note was last edited, from its live mtime) and a stale boolean (true past ~1 year) — use these to flag a recalled fact as possibly out-of-date instead of stating it as current. Ranking stays relevance-driven by default; if the server was started with --recency-weight, fresher notes are blended upward.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryYesFree-form query — multi-word natural language is the sweet spot
folderNoRestrict to a subfolder (vault-relative)
limitNoMax hits (default 10)
min_signalsNoFilter: only return hits that appeared in at least this many ranker signals. Default 1 (any). Set to 2+ for high-precision multi-ranker consensus.
embedding_modelNoOverride the embedding model alias (default 'multilingual'). Only consulted if a .embed.db exists.
granularityNov2.2.0: 'note' (default) returns one hit per note (best chunk wins). 'block' keeps each chunk as a distinct hit — useful when one note covers a topic in multiple paragraphs and you want the LLM to see all of them.
graph_boostNov2.3.0: post-RRF wikilink graph-boost — rerank top-K by counting how many OTHER top-K hits link to each one. Default ON. Set false to disable for diagnostic comparison. The 'only enquire-mcp does this' feature: generic vector stores can't do this without an Obsidian-aware layer.
filter_frontmatterNov3.10: optional YAML-frontmatter filter — a {key: value} map. A hit is kept only if its note's frontmatter satisfies EVERY pair (AND across keys). Per key: strings match case-insensitively, an array frontmatter value matches by membership (e.g. {tags: 'project'} matches `tags: [project, x]`), and the filter value may itself be an array for OR ({type: ['meeting','decision']}). Notes with no frontmatter or missing a filtered key are excluded. Omit for no filtering (default). Filters the fused candidate pool, so a strict filter can return fewer than `limit` hits.
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description discloses extensive behavioral traits beyond annotations: auto-detection of available signals, graceful degradation, per-signal observability in results, PDF chunk handling with kind and page markers, and forgetting-aware fields (age_days, stale) for recency. Annotations declare readOnlyHint and idempotentHint, which align with the description. No contradictions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with version-labeled sections and progressive detail. However, it is relatively long, and some version-specific notes (e.g., 'v2.8.0', 'v3.10') could be streamlined without losing meaning. Every sentence provides value, but conciseness could be improved by grouping or removing redundant version markers.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (8 parameters, no output schema, multiple versioned features), the description covers all essential aspects: how the fusion works, degradation behavior, result structure (per-signal, stale flags), integration with PDFs and forgetting-awareness, and filter logic. Without an output schema, it adequately describes return values. The description is complete for both basic and advanced use cases.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

While schema coverage is 100%, the description adds rich semantics to each parameter: 'query' is described as multi-word natural language sweet spot; 'min_signals' explains its precision tuning role; 'granularity' differentiates note vs block; 'graph_boost' explains the Obsidian-aware reranking feature; 'filter_frontmatter' provides comprehensive details on matching logic. The description makes every parameter's usage and behavior clear.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly identifies the tool as the default hybrid search for v2.0, listing the specific rankers (BM25, TF-IDF, embeddings) and fusion method (RRF). It distinguishes from sibling tools by explicitly stating 'Use this instead of the individual _search_text / _full_text_search / _semantic_search / _embeddings_search tools' and mentioning diagnostic uses of those alternative tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly states when to use this tool ('default search tool') and when not to ('unless you specifically need single-ranker output for diagnostics'). It also notes graceful degradation across different signals. However, it does not address when other sibling tools like obsidian_find_similar or obsidian_dataview_query might be preferred, leaving some ambiguity for non-search tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/oomkapwn/enquire-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server