chunking-embeddings

Chunking, embeddings, and RAG pipeline integration

Quality

58%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.ai-rulez/skills/chunking-embeddings/SKILL.md

Quality

Content

68%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 a tight, high-signal reference for the chunking/embeddings subsystem, strong on conciseness and actionability with exact symbols and load-bearing gotchas. Its main weakness is workflow clarity, since it reads as API reference rather than a sequenced pipeline with validation checkpoints.

Suggestions

Add a short sequenced pipeline section (e.g., 1. configure ChunkingConfig, 2. chunk_text/chunk_for_rag, 3. generate embeddings per chunk, 4. verify normalize=true) with explicit checkpoints to raise workflow_clarity from 3.

Include one minimal runnable snippet tying chunk + embed together so the guidance is fully copy-paste ready rather than signature-only.

Consider splitting the preset table and feature-gating detail into a reference file linked from SKILL.md to strengthen progressive disclosure toward 5.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes Claude's competence with Rust/serde/ONNX, with nearly every line delivering load-bearing, non-obvious detail (serde wire-name renames, disagreeing defaults, feature gating); only minor explanatory phrasing could be trimmed, matching the score-4 anchor.	4 / 5
Actionability	It supplies concrete entry points (exact function signatures, crate paths, serde wire names, a preset table, feature flags) plus explicit "Critical Rules", giving mostly executable guidance with only minor gaps versus full copy-paste examples.	4 / 5
Workflow Clarity	Ordering is only implicit ("Chunk before embedding") via the Critical Rules with no explicit sequenced workflow or validation checkpoints, fitting the score-3 anchor of steps present but checkpoints missing; the destructive/batch cap does not apply since chunking/embeddings are non-destructive.	3 / 5
Progressive Disclosure	Content is well organized under clear headers (Chunking, Embeddings, Critical Rules, Related Skills) with one-level-deep cross-skill links signaled in Related Skills and no nested references; being slightly over 50 lines keeps it at 4 rather than 5.	4 / 5
	Total	15 / 20 Passed

Description

48%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 clearly identifies a specific technical domain but is noun-only with no explicit "when to use" guidance, leaving it mid-range on completeness and trigger coverage. It is reasonably distinct from unrelated skills yet would benefit from action verbs and explicit trigger phrasing.

Suggestions

Add concrete verbs/actions, e.g. "Splits text into chunks, generates ONNX/static embeddings, and wires them into RAG pipelines" to lift specificity from 2 toward 4.

Append an explicit "Use when..." clause naming trigger phrases (e.g., "Use when building RAG pipelines, generating vector embeddings, or splitting text for retrieval") to move completeness from 3 to 5.

Include synonyms users naturally say ("vector", "retrieval", "text splitting", "semantic search") to broaden trigger-term coverage beyond the current three keywords.

Dimension	Reasoning	Score
Specificity	The description names a concrete domain ("Chunking, embeddings, and RAG pipeline integration") but lists only noun phrases with no verbs or concrete actions, matching the score-2 anchor that names the domain with minimal actions rather than the score-3 anchor which requires 1-2 named actions.	2 / 5
Completeness	It gives a clear "what" (chunking, embeddings, RAG pipeline integration) but provides no "Use when..." clause or explicit trigger guidance, so per the missing-trigger cap it cannot exceed 3.	3 / 5
Trigger Term Quality	It surfaces three relevant, natural keywords ("chunking", "embeddings", "RAG") that users would say, but omits common variations/synonyms such as "vector", "retrieval", "text splitting", or "semantic search", fitting the score-3 anchor.	3 / 5
Distinctiveness Conflict Risk	The RAG/embedding-chunking niche is mostly distinct with specific trigger terms, carrying only minor overlap risk against a broader generic "RAG" or "embeddings" skill, matching the score-4 anchor.	4 / 5
	Total	12 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: xberg-io/xberg
Path: .ai-rulez/skills/chunking-embeddings/SKILL.md
Commit: 04336bd

Reviewed: 5 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.