model-selection

Guides AI model choices in new workflows, existing workflows, pasted workflow JSON, and model-availability questions. Load before recommending, selecting, replacing, or judging a model. Preserve working or explicitly requested IDs. An unfamiliar name or generic 404 does not establish that a model is invalid. Do not claim otherwise or suggest a replacement without provider evidence. Prefer credential-specific resource lookup when available. Use searchModels only to choose an unspecified model without a relevant credential or suitable named recommendation, never to validate a supplied ID or diagnose an existing failure. For a new choice, do not narrow discovery to a remembered model ID. Do not choose GPT-4-family or Claude 3.x for an unspecified model unless verified access constraints require them. Also activated when inspecting a model-bearing node.

Quality

85%

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

Quality

Content

78%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.

A well-crafted, lean behavioral policy: it gives a concrete precedence-ordered decision procedure with exact tool calls and careful evidence-scoping rules, organized in clean self-contained sections. The main improvement opportunities are removing the small amount of duplicated guidance and adding a concrete credential-check step plus a post-choice verification step.

Suggestions

Deduplicate repeated rules (the @default caveat and the remembered-ID warning each appear twice) to tighten the token budget.

Replace 'check relevant credentials, including Gateway credits' with the concrete command or lookup used to check them.

Add a short final step to verify the chosen model is accepted (e.g., by confirming it resolves in the workflow) to close the selection loop.

Dimension	Reasoning	Score
Conciseness	The body is dense, imperative policy with no padding or explanation of concepts Claude already knows ('Do not question its validity', 'A generic 404 ... does not alone prove global model nonexistence'). However, a few points are stated twice — 'Do not treat @default as a builder-hint recommendation' appears in both the intro and step 3, and the remembered-ID warning appears in both the intro and the search guidance — so it is not perfectly lean.	4 / 5
Actionability	Guidance is highly concrete, including exact tool calls with parameters ('searchModels({ provider: "openrouter", query: "claude" })', 'nodes(action="explore-resources")') and precise namespace rules for OpenRouter IDs. Minor gaps remain, e.g., 'check relevant credentials, including Gateway credits' gives no concrete command, so it falls just short of the fully executable level-5 anchor.	4 / 5
Workflow Clarity	'Choose the source' is an explicit precedence-ordered 4-step procedure (preserve existing model → credential lookup → @builderHint → searchModels), and the 'Diagnose availability' section distinguishes actual failure evidence from guesses. This is not a destructive or batch skill, so no validation cap applies, but there is no explicit verify step after a model is chosen, keeping it below the level-5 anchor with feedback loops.	4 / 5
Progressive Disclosure	The skill is fully self-contained with no bundle files (references/, scripts/, and assets/ are absent), clear section headers, and no nested references. All content legitimately belongs in the single file, and navigation is easy, satisfying the well-organized-structure standard for a reference-free skill.	5 / 5
	Total	17 / 20 Passed

Description

87%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: it states concrete capabilities, explicit load triggers, and a distinct niche with negligible conflict risk. Its only weaknesses are a prohibition-heavy framing that blurs the capability list and a few missing natural trigger synonyms.

Suggestions

Rebalance the description toward positive capabilities (e.g., 'Selects models via credential lookups, builder hints, and catalog search') so the 'what' reads as actions rather than a list of prohibitions.

Add natural user phrasings such as 'pick a model', 'which model should I use', or 'model not found' to the trigger terms.

Dimension	Reasoning	Score
Specificity	The description lists many concrete actions ('Guides AI model choices', 'Preserve working or explicitly requested IDs', 'Prefer credential-specific resource lookup', 'Use searchModels only to choose an unspecified model'), but a substantial portion of the coverage is phrased as prohibitions ('Do not claim otherwise', 'do not narrow discovery to a remembered model ID') rather than positive capabilities, leaving minor gaps.	4 / 5
Completeness	It clearly answers both questions: 'what' via 'Guides AI model choices in new workflows, existing workflows, pasted workflow JSON, and model-availability questions' and 'when' via 'Load before recommending, selecting, replacing, or judging a model' plus 'Also activated when inspecting a model-bearing node'. This matches the level-5 anchor with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural trigger terms are well covered ('recommending, selecting, replacing, or judging a model', 'model-availability questions', 'model-bearing node', 'GPT-4-family or Claude 3.x'), but common user phrasings such as 'pick a model', 'which model', or 'model not found' are absent. Coverage is genuinely good but a few natural synonyms are missing, matching the level-4 anchor.	4 / 5
Distinctiveness Conflict Risk	The skill occupies a clear niche (model selection and validation within workflow nodes) with distinct triggers (searchModels, workflow JSON, generic 404, model-bearing node), making confusion with other skills unlikely.	5 / 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: n8n-io/n8n
Path: packages/@n8n/instance-ai/skills/model-selection/SKILL.md
Commit: 439e8ba

Reviewed: 1 day 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.