hybrid-search-implementation

Combine vector and keyword search for improved retrieval. Use when implementing RAG systems, building search engines, or when neither approach alone provides sufficient recall.

1.13x

Quality

—

Does it follow best practices?

Impact

93%

1.13x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with several complete, executable implementations and clear sectioning. Its weaknesses are the absence of validation/feedback loops in the pipelines and the lack of progressive disclosure, since all four large templates are inlined in a single file with no reference files.

Suggestions

Move the PostgreSQL and Elasticsearch engine-specific templates into separate reference files (e.g., references/postgres.md, references/elasticsearch.md) and link to them from a concise overview to improve progressive disclosure.

Add explicit validation/error-recovery steps to the RAG pipeline — e.g., check for empty vector or keyword results before fusion and fall back to the non-empty side — to add feedback loops.

Tighten the inlined code by extracting the shared RRF/linear fusion helpers once and referencing them, rather than re-implementing fusion logic inside each engine template.

Dimension	Reasoning	Score
Conciseness	The body is not padded with concepts Claude already knows, but four full executable templates (RRF, PostgreSQL, Elasticsearch, custom RAG pipeline) are inlined in a single file when some could be split out, so it is mostly efficient yet could be tightened.	2 / 3
Actionability	It provides multiple complete, executable Python implementations — RRF fusion, a pgvector hybrid search class, an Elasticsearch hybrid search class, and a full RAG pipeline — that are copy-paste ready with concrete signatures and SQL/ES bodies.	3 / 3
Workflow Clarity	The custom RAG pipeline sequences steps explicitly (Step 1: embed, Step 2: parallel searches, Step 3: fuse, Step 4: rerank), but the templates lack validation checkpoints or error-recovery feedback loops (e.g., handling empty result sets), leaving checkpoints implicit.	2 / 3
Progressive Disclosure	Sections and templates are well organized, but everything lives in one monolithic file with no bundle files or one-level-deep references, so content that could be separate (per-engine templates) is inlined rather than split.	2 / 3
	Total	9 / 12 Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise, third-person, and clearly answers both what the skill does and when to use it, with natural trigger terms. Its only weakness is that it frames a single combined action rather than enumerating multiple concrete capabilities.

Dimension	Reasoning	Score
Specificity	It names the domain and a concrete action ("Combine vector and keyword search for improved retrieval") but only describes a single composite action rather than listing multiple distinct capabilities, so it falls short of the multiple-concrete-actions anchor.	2 / 3
Completeness	It explicitly answers both what ("Combine vector and keyword search for improved retrieval") and when ("Use when implementing RAG systems, building search engines, or when neither approach alone provides sufficient recall"), matching the explicit-trigger anchor.	3 / 3
Trigger Term Quality	The "Use when" clause surfaces natural user terms — "implementing RAG systems", "building search engines", and "neither approach alone provides sufficient recall" — which are phrases a user would plausibly say when needing this skill.	3 / 3
Distinctiveness Conflict Risk	Combining vector and keyword search is a clear niche, and the "neither approach alone provides sufficient recall" trigger is specific enough to avoid firing for a pure vector or pure RAG skill.	3 / 3
	Total	11 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (571 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: Dicklesworthstone/pi_agent_rust
Commit: e9f9b67

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.