complex-task

长程/复杂任务编排工作指导。接收复杂任务后，依次完成验收标准确认、计划拆分与自审、项目文件初始化、子任务文档创建、执行者-评估者 subagent 并行迭代、主 agent 巡检协调、聚合与整体评估，直至满足验收标准后交付。

Quality

55%

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/complex-task/SKILL.md

Quality

Content

58%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 skill offers a rigorous, well-sequenced orchestration workflow with strong validation loops and actionable templates, but it is markedly verbose and monolithic — everything lives in one large SKILL.md with no progressive disclosure or bundle files.

Suggestions

Move the large reusable templates (PROJECT.md, tasks.json schema, task.md/acceptance.md skeletons, anti-pattern tables) into separate reference files under references/ and link to them from SKILL.md to improve progressive disclosure.

Cut re-explanation of concepts Claude already knows (e.g., why constraints must be documented, what a planner role is) and consolidate the status-discipline rules repeated across Phases 3, 5, and the Anti-patterns section.

Lead SKILL.md with a concise quick-start overview of the 8 phases, then defer per-phase detail to referenced files so the overview stays lean.

Dimension	Reasoning	Score
Conciseness	The body is a 530-line monologue that re-explains orchestration concepts Claude already knows (what a planner is, why constraints matter) and restates the same discipline rules across multiple phases; substantial padding could be trimmed without losing guidance.	2 / 5
Actionability	It provides concrete, copy-paste-ready templates (PROJECT.md, tasks.json schema, task.md/acceptance.md templates, executor/evaluator output formats) and specific status-flow rules; minor gaps are that some steps describe policy rather than executable actions.	4 / 5
Workflow Clarity	Phases 1–8 are clearly sequenced with explicit gates (未经确认不得进入 Phase 3, 未通过计划自审不得进入 Phase 4) and strong validation/feedback loops (GAN executor↔evaluator cycles, overall-evaluator rework loop, status-update discipline), matching the anchor with checklists and error recovery.	5 / 5
Progressive Disclosure	All content is inlined into a single SKILL.md with no bundle files in references/, scripts/, or assets/; the large template/schema blocks (tasks.json, PROJECT.md, anti-pattern tables) clearly belong in separate reference files but are not split out or signaled.	2 / 5
	Total	13 / 20 Passed

Description

51%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 conveys a concrete multi-phase orchestration workflow but is written in internal process language rather than natural user trigger phrases and lacks an explicit 'Use when...' clause, capping completeness. It is reasonably distinct from other skills.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases a user would say (e.g., 'Use when a task spans many delivery phases, requires coordinating multiple crews, or is too long for a single conversation').

Replace internal jargon (GAN 模式, verdict, spawn) in the description with plain-language outcomes so trigger-term quality matches what users actually say.

Tighten the description to lead with the observable outcome (delivers a multi-phase complex task end-to-end) before listing the internal phases.

Dimension	Reasoning	Score
Specificity	Lists several concrete orchestration actions (验收标准确认、计划拆分与自审、子任务文档创建、并行迭代、巡检协调、聚合与整体评估), but phrasing is somewhat abstract about the mechanism rather than the concrete deliverables produced.	4 / 5
Completeness	It clearly states what the skill does (orchestration workflow through delivery), but the 'when' is only weakly implied via '长程/复杂任务编排'; there is no explicit 'Use when...' clause naming concrete trigger scenarios.	3 / 5
Trigger Term Quality	The description relies on internal process jargon (subagent 并行迭代, GAN-style reviewer, verdict) rather than natural phrases a user would say when needing this skill; it lacks plain trigger language like '复杂任务' being framed as a user utterance.	2 / 5
Distinctiveness Conflict Risk	The long-form multi-phase orchestration niche is fairly distinct from typical single-purpose skills, with only minor overlap risk against generic planning or project-management skills.	4 / 5
	Total	13 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (535 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	13 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: skills/complex-task/SKILL.md
Commit: fdc03d6

Reviewed: 11 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.