sales-cs-review

当用户想复盘或升级已启用的 sales-cs 时使用。扫描 sales-cs 的 feedback/ 聚合客户反馈，结合用户意见提出升级建议（调整 MEMORY 客服手册 / 话术 / IDENTITY 称呼 / DECLARED_SKILLS 等），确认后由 main agent 直接改 sales-cs workspace 文件。 sales-cs 是对外 crew，不自行升级，所有调整经本技能由 main agent 落地。

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%

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

The content is concise, highly actionable, and presents a clear validated workflow with well-delegated script references and explicit boundaries. Minor residual imprecision (a placeholder workspace path and slight Step 3/4 overlap) does not pull any dimension below 3.

Dimension	Reasoning	Score
Conciseness	Lean ~70-line body with no padding of concepts Claude already knows; every section (triggers, prerequisites, 4 steps, boundaries, pitfalls) earns its place and command blocks are minimal, matching the score-3 anchor of a lean, token-efficient skill.	3 / 3
Actionability	Provides a concrete executable command (python3 scan_feedback.py [--since DATE], verified as a real bundle file), a copy-paste-ready '建议改动' example, and exact target files (MEMORY.md/AGENTS.md/IDENTITY.md/SOUL.md/DECLARED_SKILLS); as an instruction skill this concrete guidance satisfies the score-3 anchor.	3 / 3
Workflow Clarity	Clear 4-step sequence (Scan → Propose → Confirm & land → optional Restart) with an explicit confirmation gate in Step 3 and a confirm-before-edit feedback loop in Pitfalls, plus explicit 可改/慎改/不改 boundaries for the customer-facing destructive context.	3 / 3
Progressive Disclosure	Short, well-sectioned skill with one one-level-deep reference; the scan logic is correctly delegated to scripts/scan_feedback.py (verified present and matching its described JSON output) rather than inlined, with no nested references.	3 / 3
	Total	12 / 12 Passed

Description

85%

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, complete, and well-scoped with explicit triggers and clear distinctiveness from sibling skills. Its main weakness is trigger-term quality: natural user-facing phrasing is framed as system narration and the most idiomatic user utterances are deferred to the body rather than surfaced in the description itself.

Suggestions

Echo natural user phrasing in the description (e.g. 'Use when the user says 复盘下 sales-cs / 看看客服最近怎么样 / 调整下客服话术') so trigger terms match what a user would actually say, lifting trigger_term_quality to 3.

Keep the existing explicit trigger clause but rephrase it from third-person system voice toward user-intent voice to raise specificity of when-to-use.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions across specific artifacts — '扫描 sales-cs 的 feedback/ 聚合客户反馈', '提出升级建议（调整 MEMORY 客服手册 / 话术 / IDENTITY 称呼 / DECLARED_SKILLS 等）', '由 main agent 直接改 sales-cs workspace 文件' — matching the score-3 anchor of listing multiple specific concrete actions.	3 / 3
Completeness	Explicitly answers both what (scan feedback, propose upgrades, main agent edits files) and when ('当用户想复盘或升级已启用的 sales-cs 时使用'), with an explicit trigger clause present so the missing-'Use when' cap at 2 does not apply.	3 / 3
Trigger Term Quality	Contains some relevant terms ('复盘', '升级', '客服手册', '话术', '称呼') but they are framed as internal narration rather than echoing natural user phrasing; common variations a user would actually say live in the body, not the description, so coverage is partial per the score-2 anchor.	2 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (sales-cs review/upgrade only) and explicitly disambiguates from the sibling sales-cs-enablement skill ('sales-cs 是对外 crew，不自行升级'), giving distinct triggers unlikely to conflict.	3 / 3
	Total	11 / 12 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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: crews/main/skills/sales-cs-review/SKILL.md
Commit: bcef823

Reviewed: 3 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.