miloco-miot-scope

感知范围控制 — 管理 miloco 感知哪些家庭、哪些摄像头，每台摄像头的声音（是否参与感知），以及每台摄像头的「感知须知」（专属 prompt，补机位环境 / 关注 / 忽略以提升感知准确性）。用户说「只用/不用某家庭」「让 miloco 感知/别感知某家庭或摄像头」「屏蔽某摄像头/把某摄像头从感知里去掉」「把某摄像头声音关了/打开」「XX 老误报（声音类）」「哪些家庭在用」时激活；也在用户吐槽**画面/视觉类固定误识**时激活——「门口摄像头老把电梯门当成我家门」「XX 老把 Y 认成 Z」「这台摄像头总把走廊的人当成家里人」「能不能告诉它忽略窗外的马路」等，用「感知须知」给该机位补指导。注意区分：开关摄像头设备本身（开机/关机/电源/录制）走 miloco-devices；感知引擎自身的开关/参数走 miloco-perception。

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.

A highly actionable, well-sequenced CLI skill with real commands, example outputs, and a confirmation-gated prompt-writing workflow. Its main weaknesses are redundancy (the camera-routing logic restated several times) and the absence of any file-level split despite a substantial single-file body.

Suggestions

Consolidate the four-way camera routing (device control vs perception scope vs audio toggle vs perception-note) into one authoritative section; it is currently restated across 「工作方式」「何时激活」「⚠️ 摄像头」and「声音类 vs 画面类」.

Move the verbose 示例 JSON envelopes into a separate references/EXAMPLES.md and keep only 1–2 representative examples inline, eliminating the repeated full-envelope output per command.

Split the disambiguation matrix (本 skill vs miloco-devices / miloco-perception / miloco-miot-admin) into a references/ROUTING.md so the SKILL.md body stays a lean overview.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and skill-specific (no generic concept explanations), but the four-way camera routing is restated across 「工作方式」「何时激活」「⚠️ 摄像头」and「声音类 vs 画面类」, and full JSON envelopes repeat per command in 示例 — it could be tightened.	2 / 3
Actionability	Provides fully executable, copy-paste-ready guidance: the complete command block («miloco-cli scope home list \| switch <id>», «scope camera mic-on <did>…») plus real example invocations with actual JSON output, not pseudocode.	3 / 3
Workflow Clarity	The 5-step 「处理画面类误识」flow has explicit checkpoints («先复述给用户确认再写», «已有须知时是追加不是覆盖…先读出旧文本»), reinforced by the 校验行为 table and «先 list 确认合法再操作» — clear sequence with feedback loops.	3 / 3
Progressive Disclosure	No bundle files exist; everything is inline in one ~150-line file. Sections are clearly headed, but the lengthy 示例 block and multi-section disambiguation could be split into reference files, and the file exceeds the simple-skill (<50 line) allowance for a 3.	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.

A strong, third-person description that specifies four concrete capabilities, supplies abundant natural user-voice triggers for both audio and visual cases, and explicitly distinguishes the skill from adjacent miloco skills. It is somewhat long, but every clause earns its place via trigger coverage and conflict avoidance.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions — 「管理 miloco 感知哪些家庭、哪些摄像头，每台摄像头的声音…每台摄像头的「感知须知」」 — enumerating four distinct capabilities rather than naming only a domain.	3 / 3
Completeness	Clearly answers both what («管理 miloco 感知哪些家庭、哪些摄像头…») and when with an explicit trigger clause («用户说…时激活»); the when is stated, not merely implied.	3 / 3
Trigger Term Quality	Excellent coverage of natural phrasings a user would actually say, e.g. 「只用/不用某家庭」「屏蔽某摄像头」「把某摄像头声音关了」「门口摄像头老把电梯门当成我家门」, spanning both audio-misreport and visual-misrecognition variants.	3 / 3
Distinctiveness Conflict Risk	Explicit disambiguation — «注意区分：开关摄像头设备本身…走 miloco-devices；感知引擎自身的开关/参数走 miloco-perception» — carves a clear niche and actively prevents overlap with sibling skills.	3 / 3
	Total	12 / 12 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
metadata_field	'metadata' should map string keys to string values	Warning

	Total	15 / 16 Passed

Repository: XiaoMi/xiaomi-miloco
Path: plugins/skills/miloco-miot-scope/SKILL.md
Commit: f156d34

Reviewed: 2 days 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.