miloco-create-task

创建/管理"持续运转的家庭任务"—— 由 rule（条件触发自动化）/ schedule（定时提醒）/ record（行为累积统计）/ lifecycle（长期/限时）组合装配；也是 [感知引擎] 语音指令类系统消息的统一入口。覆盖：定时/累积提醒、"X 时 Y" 自动化、行为统计、持续状态阈值等。

Quality

68%

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 ./plugins/skills/miloco-create-task/SKILL.md

Quality

Content

77%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 and exceptionally well-sequenced with validation checkpoints and rollback, but it is a large monolithic document that could split its examples and detailed decision trees into reference files and trim some redundancy.

Suggestions

Move the 12 worked examples into a separate references/examples.md and keep a short pointer in SKILL.md, reducing the main body to the core decision procedure.

Consolidate the repeated '[输出协议 reload]' banners into a single statement near the output protocol section instead of restating it before each layer.

Deduplicate examples 4 and 9 (both '每天喝 8 杯水') into a single example that notes the first-person vs. named-person person-list resolution difference.

Dimension	Reasoning	Score
Conciseness	The body is dense domain-specific procedure with no concept padding Claude already knows, but at ~1000 lines it carries avoidable redundancy: the '[输出协议 reload]' banner is repeated three times and examples 4 and 9 are near-duplicate '每天喝 8 杯水' cases that could be tightened.	3 / 5
Actionability	Fully executable guidance: exact miloco-cli commands with precise flags, Record content JSON templates, Rule flag mapping tables, and 12 copy-paste-ready bash worked examples covering the common cases (placeholders for runtime device IDs are justified and explicitly flagged).	5 / 5
Workflow Clarity	Clear layered sequence (op routing → 前置检查 → 第一层维度判定 → 第二层子维度 → 第三层装配映射) with an explicit 装配后自检 Y/N checklist, 装配失败回滚 rollback procedure, and 'CLI 报错后重审全命令' error-recovery feedback loop.	5 / 5
Progressive Disclosure	Well-organized with headers and a clearly-signaled one-level reference (references/crud-ops.md, confirmed to exist), but the bulk of the create path — notably the 12 full worked examples (~300 lines) and detailed sub-dimension decision trees — is inlined in SKILL.md rather than split into reference files.	3 / 5
	Total	16 / 20 Passed

Description

58%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 about the domain and component model and has a clear niche, but it lacks explicit 'Use when…' trigger guidance and leans on internal jargon that weakens natural trigger-term quality.

Suggestions

Add an explicit 'Use when…' clause with concrete user trigger phrases (e.g. 'Use when the user asks to be reminded on a schedule, track a recurring behavior count, or set up an X-then-Y home automation').

Replace internal jargon like '[感知引擎] 语音指令类系统消息的统一入口' with natural phrasings a user would actually say (提醒我, 每天记录, 没人就…).

Lead with the concrete user-facing actions (定时提醒, 累积计数提醒, 自动化触发) before the rule/schedule/record/lifecycle component model.

Dimension	Reasoning	Score
Specificity	Names the domain plus four concrete assembly blocks (rule/schedule/record/lifecycle) and lists several specific coverage areas (定时/累积提醒, 'X 时 Y' 自动化, 行为统计, 持续状态阈值), but these are capability categories rather than discrete actions, leaving minor coverage gaps.	4 / 5
Completeness	Has a clear 'what' (assemble/manage household tasks from rule/schedule/record/lifecycle) but the 'when' is only weakly implied via the 覆盖 list and the 感知引擎 entry point; no explicit 'Use when…' trigger guidance, capping completeness per the rubric.	3 / 5
Trigger Term Quality	Contains some relevant terms (提醒, 统计, 'X 时 Y' 自动化, 阈值) but mixes in technical jargon ('[感知引擎] 语音指令类系统消息的统一入口', '持续运转的家庭任务') and misses the most natural user phrasings like '提醒我每天…'.	3 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (miloco persistent household-task assembly tied to 感知引擎 voice commands) with distinct triggers; minor overlap risk with sibling skills (miloco-terminate-task, crud-ops) since '创建/管理' is broad.	4 / 5
	Total	14 / 20 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
skill_md_line_count	SKILL.md is long (998 lines); consider splitting into references/ and linking	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: XiaoMi/xiaomi-miloco
Path: plugins/skills/miloco-create-task/SKILL.md
Commit: 1bd0674

Reviewed: 5 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.