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

74%

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

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 a highly actionable, well-sequenced assembly workflow with strong validation/rollback feedback loops and a clean one-level reference for CRUD ops, docked only for length and a few repeated banners that could be tightened.

Suggestions

Consolidate the repeated '[输出协议 reload]' banner into a single stated-once rule to trim padding.

Extract the large create-path dimension tables (Rule.duration_seconds / exit_debounce 推荐值, Record content JSON) into a reference file to reduce the inlined monolith.

Consider a short worked 'full assembly end-to-end' example up front so the three-layer flow is graspable before the exhaustive sub-dimension tables.

Dimension	Reasoning	Score
Conciseness	Mostly efficient procedural density that assumes Claude's competence (no basic-concept padding), but the ~1100-line create path repeats the '[输出协议 reload]' banner four times and could tighten several redundant 装配提示/反问 restatements.	3 / 5
Actionability	Fully executable copy-paste-ready miloco-cli commands across 15 worked examples, a complete Rule flag 映射 table, and concrete Record content JSON templates covering the common cases.	5 / 5
Workflow Clarity	Clear three-layer sequence (前置检查 → 维度判定 → 装配映射) with explicit validation checkpoints: a 装配后自检 echo checklist, 重复检查 conflict handling, CLI报错后重审全命令, and 装配失败回滚 rollback — feedback loops for a batch/creating operation.	5 / 5
Progressive Disclosure	Non-create ops (list/logs/disable/enable/update) are cleanly offloaded to the real references/crud-ops.md via a one-level-deep signaled link table, but the create path itself is a large inlined monolith that could be further split.	4 / 5
	Total	17 / 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 concretely names the domain and its four assembly components and is written in third person, but it omits an explicit 'Use when...' trigger clause and leans on internal vocabulary over the natural phrases a user would actually say.

Suggestions

Add an explicit 'Use when...' clause naming plain user utterances (e.g. '每天提醒我喝水', '有人摔倒就报警', '没人在家就关灯') so the trigger guidance is concrete.

Replace or supplement the rule/schedule/record/lifecycle jargon in the lead with user-facing terms (定时提醒、累计统计、条件自动化、限时任务) to improve trigger-term quality.

Tighten the '覆盖' list into the trigger clause so 'what' and 'when' are answered together in one sentence.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions (rule/schedule/record/lifecycle 组合装配) plus a covered-behaviors list (定时/累积提醒、X时Y自动化、行为统计、持续状态阈值), with only minor gaps in coverage.	4 / 5
Completeness	The 'what' is clear (创建/管理持续运转的家庭任务 via four-component 装配) but the 'when' is only weakly implied via '[感知引擎] 语音指令类系统消息的统一入口' with no explicit 'Use when...' trigger guidance, capping completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	Contains some relevant natural terms (定时提醒、行为统计、X时Y自动化) but leans on internal jargon (rule/schedule/record/lifecycle 装配) and misses the plain phrasings users actually say (e.g. 每天提醒我喝水、有人摔倒就报警).	3 / 5
Distinctiveness Conflict Risk	The 持续运转家庭任务 niche with its four-component assembly model is mostly distinct from sibling skills, with only minor overlap risk against closely related miloco-terminate-task / miloco-cli skills.	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 (1137 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: cad239d

Reviewed: 14 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.