DrawOffer MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DRAWOFFER_API_KEY | Yes | Your DrawOffer API Key (starts with dok_). Required; create it at DrawOffer → 账号 → API Key. | |
| DRAWOFFER_BASE_URL | No | Optional base URL for the DrawOffer server. Defaults to https://www.drawoffer.com. | https://www.drawoffer.com |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_resumesA | 列出当前 DrawOffer 账号里的全部简历(ID、标题、是否默认简历、更新时间),按最近更新排序。 |
| get_resumeA | 读取一份简历的正文(Markdown)。适合用来按岗位改简历、准备面试或写求职信。 |
| list_jobsA | 列出账号里保存的岗位 JD,支持关键词搜索和分页。 |
| get_jobA | 读取一个岗位的完整 JD:职责描述、技能要求、城市、薪资、学历与经验要求和备注。 |
| create_jobA | 把一个岗位 JD 保存到 DrawOffer,并自动生成一条「待投递」记录。同一岗位已存在时返回已有记录。需要读写权限的 Key。 |
| list_applicationsB | 列出求职进度:每条记录对应一个岗位,包含当前阶段、状态、投递时间和备注。 |
| get_applicationB | 读取一条投递记录的当前阶段、状态、关联岗位与简历。 |
| create_applicationA | 为已保存的岗位新建投递记录。不填 applied_at 时记为「待投递」,填了则记为已投递、进入「筛选中」。岗位已有「待投递」记录时直接沿用并更新它;已经投递过的岗位会返回冲突。需要读写权限的 Key。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
The set splits cleanly into three resources (resumes, jobs, applications) each with obvious read tools, so an agent can pick list_/get_ variants confidently. The one soft spot is create_job vs create_application: create_job silently also creates a '待投递' application record, which overlaps with create_application and could make an agent choose the wrong entry point for logging a new opportunity.
Every tool follows the same verb_noun snake_case pattern (list_resumes, get_resume, create_job, create_application). Resource nouns are reused consistently across the set (resume/job/application), so names map predictably to the domain model.
Eight tools is well within the sweet spot and each one maps to a concrete job-seeking step (browse resumes, read JD, save JD, log/track application). No tool looks like filler or a near-duplicate.
Read coverage is solid for all three resources, but the write surface is thin: there is no way to update or delete a job, no way to advance or change an application's stage (the core action in a job tracker, despite get_application exposing 'current stage'), and resumes are read-only with no edit/upload path. Agents can read and create but will hit dead ends for the main lifecycle mutation.