CtrlK
BlogDocsLog inGet started
Tessl Logo

miloco-miot-identity-register

家庭成员身份注册主流程。**用户表达「想让系统认识/记住/录入某个人」的意图就触发本 SKILL,不要求出现"注册/登记"字面词**——包括上传图/视频指明某人是谁、回应 `[感知引擎]` 陌生人推送、说"建档案/录入家人/记下样子"、指认摄像头里某人身份、从近期陌生人记录里挑人建档、撤销注册、查看某人样本。覆盖"上传图/视频"与"从摄像头挑人"两条通路,所有路径需用户确认才入库。仅改 person 行(改名/删人,不涉及样本)走 miloco-miot-identity。

73

Quality

92%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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

Highly actionable, well-sequenced content with strong validation gates, error-recovery loops, and sound one-level-deep reference splitting. Its main weakness is redundancy across the trigger-routing tables and repeated constraint explanations, which inflates token cost without adding clarity.

Suggestions

Consolidate the 何时激活 routing table and the 第一步 解析触发源 table into a single source of truth to eliminate overlapping entry-form descriptions.

State each 约束 once in the 关键约束 section and have 第三步/第四步 reference it by number instead of restating the full constraint text in every step's 硬约束 block.

DimensionReasoningScore

Conciseness

Content is task-specific operational knowledge (CLI flags, return-field structures, verbatim话术) with no concept-overexplanation, but it carries genuine redundancy — the 何时激活 routing table and 第一步 table overlap, and 约束1/硬约束 are restated across 第三步, 第四步, and 关键约束. Could be tightened without losing clarity.

2 / 3

Actionability

Fully executable copy-paste commands ('miloco-cli identity register preview --images /tmp/<uuid>_1.jpg --topk 5 --save-montage ... --pretty'), exact return-field tables, and verbatim user-facing话术 templates leave no ambiguity about what to run or say.

3 / 3

Workflow Clarity

Clear 5-step pipeline with an explicit validation gate (candidate display → user confirmation → 入库, 约束1), per-step 硬约束 checklists, and a comprehensive 异常处理 table providing error-recovery话术 and retry/pagination loops. Validation is present for the batch/destructive-adjacent register operation, so it is not capped at 2.

3 / 3

Progressive Disclosure

Core workflow stays inline while detail is split into three real, one-level-deep bundle files — [references/name-parsing.md], [references/examples.md], [references/extras.md] (all verified present) — each clearly signaled with markdown links and explicit 'when to read' guidance.

3 / 3

Total

11

/

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.

A dense, third-person description that names concrete actions, surfaces natural trigger phrasings, and explicitly routes boundary cases away. It is somewhat verbose but the length is spent on task-specific triggers rather than fluff.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — '上传图/视频指明某人是谁、回应 [感知引擎] 陌生人推送、从近期陌生人记录里挑人建档、撤销注册、查看某人样本' — rather than vague language. Each named action maps to a distinct operation, matching the 'multiple specific concrete actions' anchor.

3 / 3

Completeness

Answers both 'what' (家庭成员身份注册主流程, two pathways, user confirmation required) and 'when' (explicit trigger intent list), satisfying the 'clearly answers both what AND when with explicit triggers' anchor.

3 / 3

Trigger Term Quality

Rich natural phrasings a user would actually say — '想让系统认识/记住/录入某个人', '建档案/录入家人/记下样子', '撤销注册', '查看某人样本' — with explicit note that '注册/登记' literal words are NOT required. Covers common variations well.

3 / 3

Distinctiveness Conflict Risk

Clear niche (identity registration involving samples) with explicit boundary — '仅改 person 行(改名/删人,不涉及样本)走 miloco-miot-identity' — making it unlikely to trigger for the wrong sibling skill.

3 / 3

Total

12

/

12

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (570 lines); consider splitting into references/ and linking

Warning

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 1 missing

Warning

Total

13

/

16

Passed

Repository
XiaoMi/xiaomi-miloco
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.