intent-recognition

意图识别,理解用户输入的真实意图和上下文

Quality

51%

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 ./skills/intent-recognition/SKILL.md

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.

The body is a well-structured, concise prompt template that clearly tells Claude what to analyze and exactly how to format the result. Its main weakness is actionability detail in the vaguer output fields (suggested_actions, implicit-intent evidence). Adding brief concrete guidance for those fields would round it out.

Suggestions

Tighten the vaguer output fields: give 1-2 concrete examples of valid 'suggested_actions' entries and state what counts as sufficient evidence for '隐性意图'.

Add a short note on confidence calibration (e.g. lower confidence for implicit intents) so the numeric scores are interpretable and consistent.

Clarify how 'emotional_state' and 'execution_capacity' numeric fields should be estimated, since they currently have no anchoring guidance.

Dimension	Reasoning	Score
Conciseness	The body is lean: a task statement, input placeholders, five labeled dimensions, and a JSON output schema, with almost no padding or explanation of concepts Claude already knows. It is efficient with only minor trimmable detail (the fully enumerated JSON keys), fitting anchor 4 better than the sparse anchor 5.	4 / 5
Actionability	As an instruction-only skill it gives concrete guidance: five named analysis dimensions and a precise JSON output schema to fill. Minor gaps remain (e.g. 'suggested_actions' is described only as '基于意图识别的建议行动' and how to evidence '隐性意图' is left abstract), keeping it at anchor 4 rather than fully executable anchor 5.	4 / 5
Workflow Clarity	For a single-pass analysis skill the process is unambiguous: read the input block, evaluate the five listed dimensions, then emit the specified JSON. This matches the simple-skill clarity expected at anchor 4; it lacks explicit sequencing/checkpoints, which would be needed for a 5.	4 / 5
Progressive Disclosure	The skill is under 50 lines, single-purpose, has no external references, and is organized into clear labeled sections (任务, 输入, 识别维度, 输出格式, 注意事项). Per the simple-skill exception, this well-organized structure earns anchor 5 with no bundle files needed.	5 / 5
	Total	17 / 20 Passed

Description

25%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 a terse Chinese label that names the domain but gives only a generic action and no usage triggers, making it indistinct from other agent skills. It reads more like a category name than a skill description. Rewriting it with concrete actions and an explicit 'Use when...' clause would substantially improve all dimensions.

Suggestions

Add concrete, third-person actions instead of a single generic verb, e.g. 'Classifies user input into intent types, infers implicit needs, and assesses emotional state and urgency.'

Append an explicit trigger clause such as 'Use when the user's request is ambiguous, multi-intent, or lacks a clear goal.'

Include natural trigger terms and synonyms a user might actually say (e.g. 'what does the user really want', 'ambiguous request', 'hidden intent') to improve distinctiveness and trigger-term quality.

Dimension	Reasoning	Score
Specificity	The phrase names the domain ('意图识别') but the only action ('理解用户输入的真实意图和上下文') is minimal and generic, lacking concrete capabilities. It fits anchor 2 (names domain, actions minimal/generic) better than 3, which requires 1-2 concrete actions.	2 / 5
Completeness	It offers a vague 'what' (recognize/understand intent) with no 'when' / 'Use when...' trigger clause at all. Per the judging guidelines, a missing explicit trigger guidance caps completeness at 3, and the vague-what-plus-no-when pattern matches anchor 2.	2 / 5
Trigger Term Quality	'意图识别' is technical jargon and '理解...真实意图' is generic; there are no natural phrases a user would actually say when needing this skill. It is marginally above the no-keywords anchor (1) but misses the natural phrasing expected at 3.	2 / 5
Distinctiveness Conflict Risk	'理解用户输入的真实意图和上下文' is very broad and would overlap with many conversational or agent skills, matching anchor 2 (very broad, high overlap risk) rather than the somewhat-specific anchor 3.	2 / 5
	Total	8 / 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: ooooooooooooooooooop/ai_life_os
Path: skills/intent-recognition/SKILL.md
Commit: 61ead80

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.