Comprehensive Model Context Protocol server for managing Home Assistant through AI assistants.
Nothing to run from here
Career Kit is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
97/100
trust score
26
downloads a day
local
runs on your machine
117
downloads in 30d
What it does
AI career coach: real job-market gap analysis, staged roadmap, check-in tasks, application funnel.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user career-kit-mcp '--' uvx career-kit
# 2 — ask your agent for something these tools do
# research_market, gather_research_data, read_research_details and 38 more
Career Kit costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
W7All 1 server
w7uuuuz1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Career Kit as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need uv (docs.astral.sh/uv): uvx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user career-kit-mcp '--' uvx career-kitReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
Nothing is required, but it can take CAREER_KIT_DATA_DIR (Override the program data directory (default: ~/.career-kit)), CAREER_KIT_ASSETS_DIR (Override the generated assets directory (default: ~/CareerKit)) if you want to set them.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
41 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
市场调研：建档后的独立一步——批量抓取真实 JD 与同学学历经历。
Takes no parameters.
批量抓取真实市场数据并生成调研报告（调研步核心动作，一次调用完成）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| jd_specs | string | no | JD 抓取范围 JSON 数组，目标 1 家 + 对照 2-4 家，如 [{"company": "字节", "limit": 6}, {"company": "boss", "city": "北京", "limit": 6}]； company 支持别名（字节/boss/官方源id），其余键原样透传给对应源（city/keyword/limit…） |
| case_query | string | no | 同学历案例查询 JSON 对象，如 {"role": "AI Agent 开发", "education": "双非本", "company": "字节", "kind": "both"} ——education 用五档四层写法（985/211/双非/专科 × 博/硕/本/专） |
| direction | string | no | 报告标题与档案归档使用的方向名（默认取 case_query.role） |
| city | string | no | 报告上下文城市（默认取 case_query.city / want / who） |
| boss_confirmed | boolean | no | 范围含 BOSS 直聘时必须先问用户，用户同意后传 true。 BOSS 需要扫码登录（登录态 2 小时有效）；未确认会返回 CONFIRMATION_REQUIRED 让你先去问用户；确认但未登录会在抓取前直接返回 LOGIN_REQUIRED（不浪费整批抓取）， 让你问用户是否现在登录。用户同意后用 harness 的终端/Bash 直接运行 login_command（用户只需扫码），不要把 |
按花名册标识读取调研详情：先筛选、后读取，避免一次性加载全量正文。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| ids | string | no | 可选，JSON 数组或逗号/空格分隔的标识（如 "JD01,JD03"）； 不传时返回花名册（标识与摘要）及聚合摘要 |
| kind | string | no | "jd"（岗位，默认）/ "case"（面经与他人发展路径） |
| limit | integer | no | 单次读取上限（默认 6，最大 10）；建议分批读取 |
从路线图生成当前阶段的任务列表。
Takes no parameters.
生成当前阶段的详细路线（执行期细化）。
Takes no parameters.
保存当前阶段的详细路线。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| detail_json | string | yes | 详细路线的 JSON 字符串 |
获取当前阶段的下一步任务。
Takes no parameters.
打卡任务。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| task_id | string | yes | 任务 ID（从 get_next_tasks 获取） |
| status | string | no | 打卡状态（completed 或 skipped） |
| notes | string | no | 备注（可选） |
| amount | number | no | 打卡进度（可选）——daily 传 1 表示今天打卡（同日幂等）， 传 >1 表示目标天数；percent 传目标百分比（绝对值，如 60 表示 60%） |
触发洞察检查。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| trigger_type | string | no | 触发类型（stage_audit / event） |
| event_description | string | no | 事件描述（当 trigger_type 为 event 时必填） |
应用洞察分析结果到档案。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| insight_json | string | yes | LLM 返回的洞察分析 JSON |
获取进度概览。
Takes no parameters.
周检视——周期性事实复盘（打卡/阶段/求职漏斗）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| days | integer | no | 回看天数（1-90，默认 7；月度复盘传 30） |
获取当前工作流状态和下一步建议。
Takes no parameters.
生成自包含 HTML（双击即可在浏览器查看，无需服务器）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| mode | string | no | 视图模式 - "progress"（默认）：阶段进度仪表盘——总体/阶段进度、当前任务、能力证据、调整历史 - "roadmap"：职业地图——完整路线图（阶段/KPI/里程碑/任务/完成标准/简历价值/ 目标公司/推荐理由/JD 依据与占位徽标）+ 每阶段执行进度 + 当前阶段高亮； 执行中随时重新生成，永远是「最新地图」 |
目标选择——用户没有明确职业方向时，通过对话探索并选定目标。
Takes no parameters.
对比现状（have）与目标（want/target_jd），输出差距分析。
Takes no parameters.
保存差距分析结果。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| gap_json | string | yes | 差距分析的 JSON 字符串 |
基于差距分析生成分阶段职业路线图。
Takes no parameters.
保存路线图到档案（只校验结构；质量项以建议形式返回，不阻断保存）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| roadmap_json | string | yes | 路线图的 JSON 字符串 |
| force | boolean | no | 已废弃，保留兼容（不再影响保存） |
保存路线图审计结果（第二意见；绑定当前路线图版本，不阻断定稿）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| audit_json | string | yes | 审计结果的 JSON 字符串 |
初始化新的职业规划会话。
Takes no parameters.
解析简历文件，提取文本内容。支持 PDF、DOCX、Markdown、TXT 格式。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| file_path | string | yes | 简历文件的绝对路径 |
逐步填充档案信息。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| section | string | yes | 填充到哪个 section，只接受 who / have / want - who: 你是谁（姓名、教育、状态） - have: 你有什么（技能、经历、资源）；关键技能请附带证据与置信度， 并标 authorship（作者身份）+ control_level（把控度自评 0-10）， 整份简历自评写入 have.self_assessment - want: 你想要什么（目标岗位、行业、薪资） - |
| data | string | yes | 用户提供的信息，必须是 JSON 对象字符串 示例：'{"name":"张三", "education":"计算机本科", "skills":["Python", "React"]}' 非法 JSON 会被拒绝且不写入档案 |
确认档案信息完整，生成摘要，解锁分析工具。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| skip_probing | boolean | no | 是否跳过「简历技能摸排」（默认 False）。 摸排是硬性必选步骤：have 有技能但无用户确认证据 （user_verified）或缺 authorship/control_level/self_assessment 自评时， 本工具拒绝放行，除非用户明确表示跳过，此时 LLM 传 skip_probing=True 并记录留痕——不得以置信度高或简历字面为由自动跳过， 也不得由 AI 代填 a |
导入目标岗位的 JD（职位描述）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| jd_text | string | yes | JD 的结构化 JSON 文本 示例：'{"company":"字节跳动","role":"AI Agent 工程师","requirements":["LangGraph","RAG"]}' 空串 / 非 JSON / 非对象输入会被拒绝，不会污染档案。 |
从文件导入目标岗位的 JD（职位描述）。支持 PDF、DOCX、Markdown、TXT 格式。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| file_path | string | yes | JD 文件的绝对路径 |
记录或更新求职投递进展（求职漏斗）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| company | string | no | 公司名称（新建时必填；更新时可省略，与记录不符时以记录为准） |
| role | string | no | 岗位名称 |
| status | string | no | 当前状态（applied/screening/interview/offer/rejected/withdrawn） |
| channel | string | no | 投递渠道（boss/官网/内推/其他） |
| source | string | no | JD 来源（URL 或文件名，可选） |
| notes | string | no | 备注（可选，如面试反馈、拒信原因） |
| application_id | string | no | 更新目标（app_NNN）；为空时新建记录 |
| allow_duplicate | boolean | no | — |
删除一条投递记录（录错、重复录入时使用，不可恢复）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| application_id | string | yes | 目标记录 ID（app_NNN） |
搜寻同背景成功经验与面试要求（聚合本地知识库 + 牛客面经 + GitHub 经验帖 + 目标企业 JD）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| role | string | yes | 岗位方向（如 "AI Agent 开发"、"后端"） |
| education | string | no | 学历条件，支持「双非本」「211硕」「985」「专科」等（五档四层自动归一， 展开为搜索关键词 + 精确打分；给不出想去档位就留空） |
| company | string | no | 目标公司（可选，命中官方源时按面试要求口径补 JD；"字节"/"boss" 等别名可直传） |
| city | string | no | 城市（可选，官方/BOSS 源生效） |
| kind | string | no | "both"（默认）/ "success"（只要成功经验）/ "requirements"（只要面试要求） |
| limit | integer | no | 返回条数上限（默认 12） |
检索本地知识库（用户积累的求职资料）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | yes | 搜索关键词 示例：「AI Agent 面经」「字节跳动 JD」「双非 转 AI」 |
导入已有的职业规划文档。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| file_path | string | yes | 计划文件的绝对路径 |
获取指定企业数据源的完整使用教程。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| company | string | yes | 企业 ID（通过 list_data_sources 获取） 示例：bytedance |
列出所有已注册的企业招聘数据源。
Takes no parameters.
搜索指定企业的岗位或面经（实时抓取真实数据）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| company | string | yes | 企业 ID（通过 list_data_sources 获取） 示例：bytedance |
| params | string | no | 搜索参数 JSON 字符串（各企业支持的参数不同） 示例：'{"keyword":"AI Agent", "city":"北京"}' |
获取岗位详情或面经全文（JD 完整描述、任职要求）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| url | string | yes | 岗位详情页 URL（来自 fetch_company_jobs 结果） |
| company | string | no | 企业 ID（可选，不填则自动按 URL 匹配 Scraper） |
获取轻量行业快照（扫盲子流程）。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| industry | string | no | 行业 id、名称或别名（MVP 支持 internet / 互联网 / 软件 / 计算机； energy / 能源 / 能源动力 / 电力 / 新能源； electronics / 电子 / 半导体 / 芯片 / 集成电路 / 通信） |
| major | string | no | 可选，用于展示“你的专业在行业中的位置”；专业不是入口 |
| school | string | no | 可选，当前仅记录，不用于同校过滤 |
| degree | string | no | 可选，当前仅记录 |
| graduation_year | string | no | 可选，当前仅记录 |
| city | string | no | 可选，招聘样本城市；空为全国 |
| limit | integer | no | 招聘样本上限（1-40） |
列出所有职业档案，标记当前使用中的档案。
Takes no parameters.
切换当前使用的档案。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| profile_name | string | yes | 目标档案名（用 list_profiles 查看可用档案） |
删除一份职业档案——移入回收站，可恢复。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| profile_name | string | yes | 要删除的档案名（用 list_profiles 查看可用档案） |
| confirm | string | no | 必须显式传 "true" 才会执行；其他值只返回确认提示不删除 |
列出回收站中的档案（已删除、可恢复项）。
Takes no parameters.
从回收站恢复档案。
| Parameter | Type | Required | Notes |
|---|---|---|---|
| profile_name | string | yes | 要恢复的档案名（用 list_trash 查看可恢复项） |
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Career Kit or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 97 / 100
Release 0.2.3 · scanned 7 Nov 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score97
Findings2 threat · 1 capability
Capabilitymoderate
Coveragesource
Verificationsource
Surface digest65108528c4eee618db399ff028a5b1e47fbf46dd250c1f5f465996b83912e2e9
Score vectorthreat 100 → client 97
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | moderate | capability blast radius (moderate) — client exposure if the model is manipulated | −3 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (41 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. moderate is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| explore_goals | no tags | |
| analyze_gaps | no tags | |
| save_gap_analysis | no tags | |
| record_application | no tags | |
| delete_application | no tags | |
| search_cases | no tags | |
| search_knowledge | no tags | |
| import_plan | no tags | |
| get_scraper_guide | no tags | |
| list_data_sources | no tags | |
| fetch_company_jobs | untrusted-input | untrusted-input: keyword "fetch" |
| fetch_jd_detail | untrusted-input | untrusted-input: keyword "fetch" |
| get_industry_snapshot | no tags | |
| list_profiles | no tags | |
| switch_profile | no tags | |
| delete_profile | no tags | |
| list_trash | no tags | |
| restore_profile | no tags | |
| research_market | no tags | |
| gather_research_data | no tags | |
| read_research_details | no tags | |
| generate_roadmap | no tags | |
| save_roadmap | no tags | |
| save_roadmap_audit | no tags | |
| start_session | no tags | |
| parse_resume | no tags | |
| intake | no tags | |
| finalize_profile | no tags | |
| import_jd | no tags | |
| import_jd_file | no tags | |
| generate_tasks | no tags | |
| detail_current_phase | no tags | |
| save_current_detail | no tags | |
| get_next_tasks | no tags | |
| checkin_task | no tags | |
| trigger_insight | no tags | |
| apply_insight | no tags | |
| get_progress | no tags | |
| get_weekly_review | no tags | |
| get_workflow_status | no tags | |
| export_dashboard | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest65108528c4eee618
Scanned2026-11-07
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.2.3. Pin a release below and the command asks for exactly that one.
0.2.3
7 Nov 2026
latestNo release note was published with this version.
0.2.2
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Career Kit can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.