faiss

Fast vector similarity search at billion scale.

Quality

53%

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 ./optional-skills/mlops/faiss/SKILL.md

Quality

Content

57%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 with executable examples across FAISS's main features and clear section structure, but it underuses its own reference bundle (index_types.md is orphaned), inlines content that belongs in the reference, and lacks explicit workflow/validation guidance. Tightening promotional metrics and routing detail to the reference file would improve it.

Suggestions

Replace the inlined 'Index types' section with a concise summary and a link to references/index_types.md to fix the orphaned reference and reduce duplication.

Add an explicit build workflow (choose index → train if needed → add → search → save) with a validation checkpoint such as asserting index.is_trained before add.

Trim promotional metrics ('31,700+ GitHub stars', 'Meta/Facebook AI Research') that do not aid execution.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with tight code blocks and tables, but it includes redundant promo-style metrics ('31,700+ GitHub stars', 'Meta/Facebook AI Research', 'Handles billions of vectors') and a redundant intro line that restates the description, so it could be trimmed.	3 / 5
Actionability	Provides concrete, mostly copy-paste-ready code across index types, save/load, GPU, and LangChain/LlamaIndex integrations, with only minor gaps such as the truncated LlamaIndex example (no search shown) and reused variables from the quick start.	4 / 5
Workflow Clarity	The train→add→search sequence is visible inside the IVF/PQ examples, but there is no explicit workflow framing or validation checkpoints (e.g. checking index.is_trained), matching the 'steps present but checkpoints missing/implicit' anchor.	3 / 5
Progressive Disclosure	Section headers are well organized, but the index-types material is fully inlined in the body while a references/index_types.md file exists and is never linked, fitting the 'references present but not signaled; content that should be separate is inline' anchor.	3 / 5
	Total	13 / 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 is concise and clearly states a specific niche, but it omits any 'when to use' trigger guidance and lists only a single action, which limits completeness and specificity. Adding a 'Use when...' clause with natural trigger terms would substantially improve it.

Suggestions

Add an explicit trigger clause, e.g. 'Use when building RAG, similarity search, or KNN over millions+ of vectors.'

List concrete actions (index, search, save/load, GPU acceleration) instead of a single 'search' verb to raise specificity.

Include natural trigger synonyms like 'vector search', 'KNN', 'nearest neighbors', and the library name 'FAISS'.

Dimension	Reasoning	Score
Specificity	Names the domain ('vector similarity search') with one action verb and scale modifiers ('fast', 'billion scale'), but lists no concrete capabilities beyond a single search action, matching the 'names domain but actions minimal/generic' anchor rather than the multi-action anchor 3.	2 / 5
Completeness	Provides a clear 'what' (fast vector similarity search at billion scale) but no 'Use when...' trigger clause, so per the guideline a missing 'when' caps completeness at 3.	3 / 5
Trigger Term Quality	Includes relevant natural phrases users say ('vector similarity search', 'billion scale') but misses common variations, synonyms, and library name 'FAISS', fitting the 'some relevant keywords but missing variations' anchor.	3 / 5
Distinctiveness Conflict Risk	The vector-similarity-search-at-billion-scale niche is mostly distinct with only minor overlap risk against other vector-DB skills (Chroma/Pinecone/Weaviate), matching the 'mostly distinct; minor overlap' anchor.	4 / 5
	Total	12 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	13 / 16 Passed

Repository: NousResearch/hermes-agent
Path: optional-skills/mlops/faiss/SKILL.md
Commit: 6680afb

Reviewed: about 5 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.