CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

email-ops-skill

当用户要对真实电子邮件账户做任何操作时触发:查收件箱、读信、回复、转发、撰写、发送、搜索、归档、删除。典型说法"查下我邮箱"、"xx 给我发了啥"、"check my inbox"。**不用于**:即时消息 / IM / 微信 / Slack、翻译含"邮件"二字的文本、对 .eml 文件做离线解析(→直接用 Read)。

70

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

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.

A strong, highly actionable instruction-only skill: concrete tool modes and parameters, four well-sequenced workflows with explicit user-confirmation validation for every destructive or batch operation, and a common-errors checklist. Its main weakness is moderate padding — the 'why' rationale paragraphs re-explain email irreversibility and threading mechanics Claude already knows, and the quick-reference section duplicates the tool table.

Suggestions

Trim the '第 5 步为什么必须做' and '第 6 步为什么必须带 inReplyTo' rationale paragraphs to one-line imperatives (e.g., '邮件发出不可撤回——必须先展示草稿并等待确认' and 'inReplyTo 填原邮件 messageId,references 填原邮件 references 链 + messageId,否则线程断裂'), removing the explanation of how Gmail/Outlook/Apple Mail organize conversations.

Remove the '操作速查' section's mode/action lists since they duplicate the '工具清单' table, or merge the two into one section.

In 流程三, make the per-batch checkpoint concrete by specifying the report format (e.g., '每批操作后报告成功/失败条数与剩余数量') so progress reporting is verifiable rather than implied.

DimensionReasoningScore

Conciseness

The body is mostly tight (tool table, mode-selection map, numbered workflows), but the 'why' rationale paragraphs are unnecessary — '因为邮件一旦发出就无法撤回。发错人、写错内容可能造成不可逆的尴尬甚至商业损失' and the explanation that '所有主流邮件客户端(Gmail、Outlook、Apple Mail)都依赖 Message-ID 头来组织会话' cover concepts Claude already knows. The '操作速查' section also re-lists the same 7 modes and 8 actions already tabulated at the top.

3 / 5

Actionability

Guidance is fully executable: exact tool modes with required parameters ('get-message (需要 messageId)', 'list-messages (需要 accountEmail + mailbox)'), concrete pagination values ('pageSize: 20', 'cursor + pageSize(1-50)'), specific send parameters ('需要 accountEmail、to、subject、bodyText;回复额外需要 inReplyTo、references'), and concrete forwarding construction ('subject 加 Fwd: 前缀,bodyText 包含 ---------- Forwarded message ---------- 分隔头'). Exact tool schemas are appropriately deferred via ToolSearch.

5 / 5

Workflow Clarity

Four workflows are clearly sequenced with explicit validation checkpoints: '⚠️ 用纯文本把草稿(收件人 / 主题 / 正文)展示给用户,并明确询问"是否发送",等用户回复确认' before send, '用户确认后才发送', and for batch operations '批量删除前务必让用户审核样本' plus per-batch progress reporting. A 7-item error checklist ('常见错误——你必须避免') covers the failure modes. Validation is present for every destructive/batch operation, so no cap applies.

5 / 5

Progressive Disclosure

Good structure with well-labeled sections and no bundle files to navigate; the ~120-line single file is cohesive operational guidance rather than an inlined API reference, and schema details are deferred to the tools themselves. Falls short of a 5 because the trailing '操作速查' partially duplicates the opening tool table and the doc is above the under-50-line simple-skill threshold where sections alone would suffice.

4 / 5

Total

17

/

20

Passed

Description

92%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.

A strong description: explicit trigger clause, nine enumerated concrete actions, natural user phrases in both Chinese and English, and an explicit not-for list that sharply delimits it from adjacent skills. Third-person voice is used throughout, with only minor gaps in natural trigger-term coverage.

DimensionReasoningScore

Specificity

Nine concrete actions are enumerated — '查收件箱、读信、回复、转发、撰写、发送、搜索、归档、删除' — giving comprehensive coverage of the email-account operation domain, matching the top anchor of multiple specific concrete actions with comprehensive coverage. Nothing is vague or generic.

5 / 5

Completeness

Both what and when are explicit: '当用户要对真实电子邮件账户做任何操作时触发' is a clear trigger clause, the action list is the 'what', and concrete trigger phrases are quoted. It exceeds the anchor-4 example by also adding explicit negative boundaries, so it clearly matches the top anchor.

5 / 5

Trigger Term Quality

Natural trigger phrases users would actually say are quoted in two languages: '"查下我邮箱"、"xx 给我发了啥"、"check my inbox"', plus the enumerated action verbs. A few natural variations are missing (e.g., English phrasings beyond 'check my inbox', or common requests like '帮我回一下这封邮件' / 'forward this'), so it sits at 'good keyword coverage; a few natural terms missing' rather than fully comprehensive.

4 / 5

Distinctiveness Conflict Risk

The negative boundary list ('**不用于**:即时消息 / IM / 微信 / Slack、翻译含"邮件"二字的文本、对 .eml 文件做离线解析(→直接用 Read)') preempts the most likely mis-triggers — chat apps, text translation, and offline file parsing — giving a clear niche with distinct triggers and minimal conflict risk.

5 / 5

Total

19

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

itsablabla/openloaf-web

Path

apps/server/src/ai/builtin-skills/email-ops/SKILL.md

Reviewed

Table of Contents

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.