CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

embedding-strategies

Guide to selecting and optimizing embedding models for vector search applications.

Quality

62%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/AI-Agents-Safe-Coding-Skills/skills/embedding-strategies/SKILL.md

The canonical home for this skill is embedding-strategies in administrakt0r/AI-Agents-Safe-Coding-Skills

SKILL.md
Quality
Evals
Security

Quality

Content

53%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 skill provides a genuinely useful model comparison table and mostly executable code templates, but it functions as an inlined code dump rather than an overview with progressive disclosure. Generic filler sections and a broken reference to a nonexistent playbook file further weaken its structure.

Suggestions

Split the large code templates (chunking strategies, evaluation metrics) into separate reference files under references/ and keep SKILL.md as a concise overview with well-signaled one-level-deep links.

Fix the dangling reference: 'resources/implementation-playbook.md' does not exist — either create the playbook file in a proper references/ directory or remove the pointer.

Replace the vague 'Instructions' bullets with a concrete sequenced workflow (select model from table -> chunk -> embed -> evaluate with Template 5 metrics) that includes an explicit validation checkpoint, and remove the generic 'Do not use this skill when' filler.

DimensionReasoningScore

Conciseness

The model comparison table and templates carry non-obvious domain info, but generic filler sections ('Clarify goals, constraints, and required inputs. Apply relevant best practices and validate outcomes.', 'The task is unrelated to embedding strategies') and Template 5's reimplementation of standard IR metrics (precision@k, MRR, NDCG) that Claude already knows add noticeable padding that could be trimmed.

3 / 5

Actionability

Templates 1-3 and 5 are mostly executable, copy-paste-ready code with specific model names and parameters. Minor gaps: Template 4's chunk_code is a 'pass' stub with comment placeholders, DomainEmbeddingPipeline uses 're'/'List' without imports, and embed_query's 'if "bge" in self.model.get_sentence_embedding_dimension()' checks a substring of an int, which is broken.

4 / 5

Workflow Clarity

This is a reference-style skill with a conceptual pipeline diagram but no operational sequence; the 'Instructions' bullets ('Apply relevant best practices and validate outcomes') are vague, and Template 5's evaluation code exists as validation tooling but is never integrated as a checkpoint in any workflow.

3 / 5

Progressive Disclosure

Roughly 350 lines of template code are inlined in SKILL.md that clearly belong in separate reference files, and the body's only file reference, 'resources/implementation-playbook.md', points to a directory that does not exist in the bundle (no references/, scripts/, or assets/ are present), making it a dangling reference.

2 / 5

Total

12

/

20

Passed

Description

53%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 states what the skill does in a specific niche, but it entirely lacks a 'when to use' trigger clause and misses the natural trigger terms (RAG, embeddings, semantic search) the body itself enumerates. Adding these would lift completeness and trigger term quality substantially.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when choosing embedding models for RAG, optimizing chunking, or comparing text embedding models for semantic search.'

Include natural trigger terms users would say: embeddings, RAG, semantic search, similarity search, vector database, chunking.

Enumerate a few more concrete capabilities from the body (chunking strategies, fine-tuning for domains, dimension reduction) instead of the generic 'selecting and optimizing'.

DimensionReasoningScore

Specificity

Names the domain ('embedding models', 'vector search applications') plus two actions ('selecting and optimizing'), matching anchor 3 ('Processes PDF files and extracts content') — but it omits the several specific actions the body actually covers (chunking strategies, fine-tuning, dimension reduction, multilingual handling).

3 / 5

Completeness

The 'what' is clear (selecting and optimizing embedding models for vector search), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which per the judging guidelines caps completeness at 3.

3 / 5

Trigger Term Quality

Includes relevant keywords 'embedding models' and 'vector search', but misses the natural phrases users would actually say — embeddings, RAG, semantic search, similarity search, chunking — so coverage falls short of anchor 4's 'good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

'Embedding models for vector search applications' names a concrete, specific niche comparable to anchor 4's 'Works with PDF and Word document files', with only minor overlap risk against closely related RAG or vector-database skills; it is more specific than anchor 3's broad phrasing.

4 / 5

Total

13

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

administrakt0r/AI-Agents-Safe-Coding-Skills

Path

plugins/AI-Agents-Safe-Coding-Skills/skills/embedding-strategies/SKILL.md

Revision

64adc55

Reviewed

Table of Contents

DiscoveryImplementationValidation

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.

Claim skill