CtrlK
BlogDocsLog inGet started
Tessl Logo

llm-structured-output

Get reliable JSON, enums, and typed objects from LLMs using response_format, tool_use, and schema-constrained decoding across OpenAI, Anthropic, and Google APIs.

62

Quality

73%

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-claude/skills/llm-structured-output/SKILL.md

The canonical home for this skill is llm-structured-output in administrakt0r/AI-Agents-Safe-Coding-Skills

SKILL.md
Quality
Evals
Security

Quality

Content

81%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 complete executable examples and a well-sequenced, validation-backed workflow including a retry feedback loop. Conciseness is the main weakness due to some explanatory prose that assumes less competence than Claude has.

Suggestions

Trim explanatory asides that restate what Claude already knows (e.g. the 'bare sentiment: str' contrast in step 4) to push conciseness toward 4-5.

Consider moving the per-provider schema snippets or the Edge Cases into a reference file so the SKILL.md overview stays leaner.

The Examples are strong; keep them but ensure surrounding prose does not re-explain library basics.

DimensionReasoningScore

Conciseness

Mostly efficient with concrete code and well-organized sections, but includes some over-explanatory prose Claude already knows (e.g. 'Models use these descriptions as implicit prompt instructions — a field described as… produces better results than a bare sentiment: str').

3 / 5

Actionability

Three complete, copy-paste-ready executable examples (OpenAI+Pydantic, Anthropic tool_use, TypeScript+Zod) plus concrete per-step guidance cover the common provider cases fully.

5 / 5

Workflow Clarity

A 10-step sequenced Core Workflow includes an explicit validation step (step 8) and a retry feedback loop (step 9: send error back, cap at 3), matching the anchor for clear sequence with validation and error-recovery loops.

5 / 5

Progressive Disclosure

Content is cleanly partitioned into well-signaled sections (When to Use, Core Workflow, Examples, Never Do This, Edge Cases, Best Practices) with no nested references, though some material (provider-specific details, examples) is inlined rather than split into reference files.

4 / 5

Total

17

/

20

Passed

Description

66%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 specific and well-targeted with good technical keyword coverage, but it omits an explicit 'when to use' trigger clause, which caps completeness. Adding a 'Use when…' clause with natural phrasings like 'structured output' would lift both completeness and trigger-term quality.

Suggestions

Add an explicit 'Use when…' clause, e.g. 'Use when extracting structured output (JSON, enums, typed objects) from an LLM response or when the user asks about json_mode, structured outputs, or tool_use for extraction.'

Include natural user-facing synonyms such as 'structured output' and 'JSON schema' alongside the existing API terms.

Replace the generic verb 'Get' with concrete actions like 'Extract and validate JSON, enums, and typed objects…' to push specificity toward 5.

DimensionReasoningScore

Specificity

Names the domain ('Get reliable JSON, enums, and typed objects from LLMs') and three concrete mechanisms (response_format, tool_use, schema-constrained decoding) across three providers, but the action verb is generic ('Get') leaving minor coverage gaps versus the comprehensive 5-anchor.

4 / 5

Completeness

The 'what' is clearly stated but there is no 'Use when…' or equivalent explicit trigger clause, so per the rubric guideline completeness is capped at 3.

3 / 5

Trigger Term Quality

Strong technical keyword coverage ('JSON', 'enums', 'response_format', 'tool_use', 'schema-constrained decoding', provider names) but omits common natural phrasings users say like 'structured output' or 'JSON schema'.

4 / 5

Distinctiveness Conflict Risk

The structured-output-from-LLMs niche across named providers is clearly distinct from neighboring skills, with only minor overlap risk against general LLM skills.

4 / 5

Total

15

/

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.

Validation15 / 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
Reviewed

Table of Contents

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.