chunking-strategies

Document chunking strategies for RAG systems. Use when implementing document processing pipelines to determine optimal chunking approaches based on document type and retrieval requirements.

Quality

75%

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 ./plugins/rag-cag/skills/chunking-strategies/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 delivers concrete, mostly-executable chunking implementations plus a useful document-type selection table in a lean, well-structured form. Its main weaknesses are the lack of a sequenced validation workflow and an undefined helper function that slightly undercuts copy-paste actionability.

Suggestions

Define or import the missing count_tokens() helper (and the dataclass import) so the semantic chunking and EnrichedChunk examples run as-is, or note the dependency explicitly.

Add a short numbered workflow (choose strategy by document type -> chunk -> validate retrieval quality -> tune size/overlap) with an explicit validation checkpoint to raise workflow clarity above the batch-operation cap of 3.

Trim the redundant intro line that restates the frontmatter description, and tighten the generic Best Practices list to skill-specific guidance.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean code with minimal prose and assumes Claude's competence; minor redundancy (the intro line restates the frontmatter and the best-practices list is somewhat generic) keeps it just below a 5.	4 / 5
Actionability	Provides mostly executable Python functions for each chunking method and a concrete selection table, but the undefined count_tokens() helper and missing dataclass import/usage examples prevent a fully copy-paste-ready 5.	4 / 5
Workflow Clarity	The content is organized as a strategy catalog rather than a sequenced workflow, and while 'test retrieval quality' is mentioned as a best practice, there are no explicit validation checkpoints; the batch-operation cap at 3 applies.	3 / 5
Progressive Disclosure	Sections are clearly headed and logically organized with content appropriately kept inline for a moderate-length strategy reference, but there are no external references at all so it does not reach the navigation-rich 5 anchor.	4 / 5
	Total	15 / 20 Passed

Description

70%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 communicates a focused niche (RAG document chunking) with an explicit Use-when trigger and concrete natural keywords. It is well above the vague minimum but stops short of the top anchor due to limited action variety and missing trigger synonyms.

Dimension	Reasoning	Score
Specificity	Names the domain and a couple of concrete actions ('determine optimal chunking approaches', 'implementing document processing pipelines') but the coverage is not comprehensive, matching the anchor for 1-2 concrete actions rather than several.	3 / 5
Completeness	Explicitly states what the skill does and includes a clear 'Use when...' clause with concrete scenario triggers (document type, retrieval requirements), but lacks the varied 'when the user mentions X' trigger-phrase list of the 5 anchor.	4 / 5
Trigger Term Quality	Includes natural terms a user would say ('chunking', 'RAG', 'document processing', 'retrieval') with good coverage, but misses common synonyms like 'splitting'/'segmentation' and any file extensions, so it falls just short of a 5.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche (chunking strategies for RAG) with distinct triggers, though there is minor overlap risk with broader RAG or document-processing skills, keeping it below 5.	4 / 5
	Total	15 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: jpoutrin/product-forge
Path: plugins/rag-cag/skills/chunking-strategies/SKILL.md
Commit: 0ebe7ae

Reviewed: 7 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.