faiss

Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.

Quality

71%

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 ./backend/cli/skills/llm-tools/faiss/SKILL.md

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 body is code-rich and largely executable, but it duplicates the bundled index-types reference inline without ever linking to it, and lacks an explicit usage workflow with validation. Tightening the marketing fluff and routing index detail to the reference file would materially improve it.

Suggestions

Link references/index_types.md from the 'Index types' section and keep only a brief selection summary inline instead of duplicating all four index-type code blocks.

Replace the inlined index-type sections with a short decision table plus a pointer to the reference, and remove the duplicated GitHub-stars/marketing metrics.

Add an explicit end-to-end workflow (choose index → create → train if needed → add → search → save) with a verification step confirming recall or index integrity.

Dimension	Reasoning	Score
Conciseness	Code examples are lean, but marketing metrics ('31,700+ GitHub stars' stated twice, 'Handles billions of vectors') and four inlined index-type sections that duplicate references/index_types.md add noticeable padding.	3 / 5
Actionability	Core usage provides fully executable, copy-paste-ready code; minor gaps remain in integration snippets (undefined 'docs' in the LangChain example, partial LlamaIndex snippet with no ingestion/search).	4 / 5
Workflow Clarity	An implicit create→train→add→search sequence is embedded in the examples, but there is no explicit sequenced workflow and no validation checkpoints; 'Best practices' is guidance rather than a workflow.	3 / 5
Progressive Disclosure	The bundle file references/index_types.md exists but is never linked or signaled in the body, while its content (Flat/IVF/HNSW/PQ) is inlined — an orphaned reference plus content that belongs in a separate file.	2 / 5
	Total	12 / 20 Passed

Description

88%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.

A strong description that clearly states both capability and explicit usage triggers with concrete, natural terms. The main gap is mild overlap with general vector-store tools and a few missing synonyms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities — similarity search, clustering of dense vectors, GPU acceleration, index types (Flat, IVF, HNSW), k-NN search, and large-scale vector retrieval — giving comprehensive coverage of what FAISS does.	5 / 5
Completeness	Explicitly answers both 'what' (library for efficient similarity search and clustering) and 'when' ('Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata').	5 / 5
Trigger Term Quality	Good natural keyword coverage ('similarity search', 'k-NN search', 'vector retrieval', 'GPU acceleration'), but missing common synonyms like 'nearest neighbors'/'ANN' and any file-extension triggers.	4 / 5
Distinctiveness Conflict Risk	The 'pure similarity search without metadata' carve-out establishes a distinct niche, but 'similarity search'/'vector retrieval' still overlaps with closely related vector-store skills (Chroma, Pinecone, Weaviate, Annoy).	4 / 5
	Total	18 / 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: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/faiss/SKILL.md
Commit: fcd2e11

Reviewed: 2 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.