using-llm

List available large language models and send chat completion requests programmatically. Use this skill when you need to call an LLM within a snippet, including model comparison, visual understanding, batch inference, and model performance testing.

Quality

88%

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

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with complete executable examples and a clear quick-start workflow, but loses points on conciseness from bilingual duplication and on progressive disclosure because the inline API reference is not split into separate reference files.

Suggestions

Remove or relocate the parallel  Chinese blocks (or move translation to a separate locale file) to roughly halve token overhead in the loaded SKILL.md.

Extract the Parameter Reference, thinking parameter table, and Return Value sections into a separate references/ file (e.g. REFERENCE.md) and link to it from the body, since the skill is well over 50 lines.

De-duplicate the image_to_base64 data-URL warning and the 'thinking disabled is recommended default' note so each appears once.

Dimension	Reasoning	Score
Conciseness	The body duplicates every English section with a parallel Chinese <!--zh--> block, roughly doubling tokens for the same information, plus repeats the image_to_base64 warning and thinking-default guidance; not a wall of background prose, but it could be tightened considerably.	2 / 3
Actionability	Provides complete, executable run_python_snippet code blocks for listing models, chatting, and vision, plus concrete parameter tables and return-value access patterns — copy-paste ready with real imports and arguments.	3 / 3
Workflow Clarity	Quick Start sequences Step 1 (list models) -> Step 2 (chat) with rationale to query first, includes a vision fallback (file_to_url -> image_to_base64) and explicit error-avoidance guardrails; the operation is read-only so the destructive-validation cap does not apply.	3 / 3
Progressive Disclosure	No references/scripts/assets bundles exist; the ~313-line SKILL.md holds the entire API/parameter and return-value reference inline, which is content the score-2 anchor flags as 'should be separate' rather than a well-signaled one-level split.	2 / 3
	Total	10 / 12 Passed

Description

100%

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 strong: it names concrete actions, provides an explicit 'Use when' trigger with natural terms, and carves out a distinct niche. It would benefit only marginally from trimming the slightly technical phrasing.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — listing models, sending chat completion requests, model comparison, visual understanding, batch inference, and performance testing — matching the 'lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly states what the skill does and provides an explicit 'Use this skill when you need to call an LLM within a snippet, including...' trigger clause with enumerated use cases.	3 / 3
Trigger Term Quality	Covers natural terms a developer would say ('call an LLM', 'model comparison', 'visual understanding', 'batch inference', 'performance testing'); slightly technical with 'programmatically'/'within a snippet' but well within good coverage.	3 / 3
Distinctiveness Conflict Risk	The LLM-calling niche via the platform SDK is well-scoped with distinct triggers, making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: dtyq/magic
Path: backend/super-magic/agents/skills/using-llm/SKILL.md
Commit: 41d7ef4

Reviewed: about 13 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.