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

71%

Does it follow best practices?

Impact

93%

1.13x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/hybrid-search-implementation/SKILL.md

The canonical home for this skill is hybrid-search-implementation in wshobson/agents

Quality

Content

65%Weight 40%Scale 1-5

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, delivering four complete executable implementations with clean sectioning. Its weaknesses are token efficiency (four large inlined templates) and the absence of progressive disclosure into separate reference files.

Suggestions

Move the longer per-engine templates (Postgres, Elasticsearch, custom RAG pipeline) into files under references/ and keep only the RRF/linear core plus a one-line pointer per engine in SKILL.md.

Add a short "Tuning workflow" section with explicit validation checkpoints (e.g., measure recall@k on a held-out set, compare fusion weights, only ship when recall improves) to give the multi-step tuning process explicit feedback loops.

Dimension	Reasoning	Score
Conciseness	There is no padding of concepts Claude already knows, but roughly 400 lines of four full inlined implementations (Postgres, Elasticsearch, custom RAG pipeline) could be tightened or moved to references, fitting the mostly-efficient-but-could-be-tighter score-3 anchor; it is above score 2 (no heavily padded explanatory prose) but below score 4 (the volume of inlined code is more than minor).	3 / 5
Actionability	Four complete, executable Python templates (RRF/linear, Postgres with pgvector+FTS, Elasticsearch, custom RAG pipeline) ship with real imports, typed signatures, and working logic, matching the fully-executable copy-paste-ready score-5 anchor.	5 / 5
Workflow Clarity	Template 4 exposes a numbered Step 1-4 sequence (embed, parallel search, fuse, rerank), but the body is a template catalog with no validation checkpoints or overall guided process, fitting the steps-present-but-checkpoints-implicit score-3 anchor; it is below score 4 which requires most checkpoints present, and the destructive/batch cap is not triggered since search is non-destructive.	3 / 5
Progressive Disclosure	Sections are well organized (When to Use, Core Concepts, Templates, Best Practices, Resources) but no bundle files exist and all content is inlined in SKILL.md, with the long per-engine templates that would belong in separate reference files not split out, fitting the some-structure-but-content-should-be-separate score-3 anchor.	3 / 5
	Total	14 / 20 Passed

Description

78%Weight 40%Scale 1-5

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 strong: it concisely states what the skill does and gives explicit, natural trigger conditions covering RAG and search-engine contexts. The main gap is the absence of the most common synonym "hybrid search" and related terms like BM25 in the trigger phrasing.

Suggestions

Include the natural phrase "hybrid search" and synonyms like "BM25" or "semantic search" in the trigger clause to improve trigger-term coverage.

Add one more concrete action verb (e.g., "fuse", "rerank") to the what-clause to lift specificity from a single combining action toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	"Combine vector and keyword search for improved retrieval" names the domain and one concrete combining action, matching the score-3 anchor of 1-2 concrete actions without comprehensive coverage; it stops short of listing multiple specific actions (score 4) and is well above the single-generic-action score-2 anchor.	3 / 5
Completeness	It explicitly answers 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") with concrete trigger phrases, matching the score-5 anchor exactly.	5 / 5
Trigger Term Quality	"RAG systems, building search engines, vector and keyword search" gives good natural-term coverage a user would say, fitting score 4; it is below score 5 because common synonyms like "hybrid search", "BM25", and "semantic search" are absent, and above score 3 because it supplies several natural phrases rather than one generic keyword.	4 / 5
Distinctiveness Conflict Risk	Hybrid search is a clear, specific niche with distinct triggers, fitting the mostly-distinct score-4 anchor; it is below score 5 because broad phrasing like "building search engines" carries minor overlap risk with general search skills.	4 / 5
	Total	16 / 20 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
Path: tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/hybrid-search-implementation/SKILL.md
Commit: 6453c0b

Reviewed: about 8 hours 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.