Tencent CloudBLOCK
Tencent Cloud MCP Server, Support Remote Hosted and Local Installation
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
A suite of specialized MCP servers that help you get the most out of TencentCloud. Build cloud-native applications, manage infrastructure, and accelerate AI development and deployment wherever you use MCP.
[](LICENSE)
🌍 Languages: English | 中文
Table of Contents
- TencentCloud MCP Servers
- Table of Contents
- What is the Model Context Protocol (MCP) and how does it work with TencentCloud MCP Servers?
- Why TencentCloud MCP Servers?
- Available MCP Servers: Quick Installation
- Project Management
- Development Efficiency
- SaaS
- PaaS
- IaaS
- When to use Local vs Remote MCP Servers?
- Local MCP Servers
- Remote MCP Servers
- Remote and Local Installation & Setup
- Steps for Using Remote Hosted MCP
- Steps for Local Installation of MCP
- Installation and Setup
- Getting Started with CodeBuddy
codebuddy_mcp_settings.json- Getting Started with Cursor
.cursor/mcp.json- Getting Started with Windsurf
~/.codeium/windsurf/mcp_config.json- Getting Started with VS Code
- [
.vscode/mcp.json](#vscodemcpjso
590795f31c64OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add tdesign-mcp-server --env API_TOKEN=${API_TOKEN} --env EDGEONE_PAGES_API_TOKEN=${EDGEONE_PAGES_API_TOKEN} --env MINIPROGRAM_PRIVATE_KEY=${MINIPROGRAM_PRIVATE_KEY} --env MINIPROGRAM_PRIVATE_KEY_PATH=${MINIPROGRAM_PRIVATE_KEY_PATH} -- npx -y [email protected]{
"mcpServers": {
"tdesign-mcp-server": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"API_TOKEN": "${API_TOKEN}",
"EDGEONE_PAGES_API_TOKEN": "${EDGEONE_PAGES_API_TOKEN}",
"MINIPROGRAM_PRIVATE_KEY": "${MINIPROGRAM_PRIVATE_KEY}",
"MINIPROGRAM_PRIVATE_KEY_PATH": "${MINIPROGRAM_PRIVATE_KEY_PATH}"
}
}
}
}Exposed tools (148)
107 read · 36 write · 5 destructive. Blast radius: 5 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
CreateAutoScalingGroup | read | 创建伸缩组 |
CreateDiagnosticReports | read | 创建实例诊断报告 |
CreateSecurityGroup | write | Create a new security group |
CreateSecurityGroupPolicies | write | Create security group rules |
CreateSecurityGroupWithPolicies | write | Create a new security group with rules |
DescribeAutoScalingGroups | read | 查询伸缩组 |
DescribeDiagnosticReports | read | 查询实例诊断报告 |
DescribeImages | read | 查询腾讯云CVM镜像列表 |
DescribeInstanceTypeConfigs | read | 查询实例机型配置列表 |
DescribeInstances | read | 查询腾讯云CVM实例列表 |
DescribeRecommendZoneInstanceTypes | read | 推荐用户在相同或相近地域购买相同或相似机型 |
DescribeRegions | read | 查询腾讯云CVM支持的地域列表 |
DescribeSecurityGroupPolicies | read | Query security group rules |
DescribeSecurityGroups | read | 查询安全组列表 |
DescribeSubnets | read | 查询子网列表 |
DescribeVpcs | read | 查询VPC列表 |
DescribeZones | read | 查询腾讯云CVM支持的可用区列表 |
DisableAutoScalingGroup | read | 停用伸缩组 |
EnableAutoScalingGroup | read | 启用伸缩组 |
ExecuteScalingPolicy | read | 执行伸缩策略 |
GetCpuLoadavgData | read | 获取CPU一分钟平均负载指标数据 |
GetCpuUsageData | read | 获取CPU利用率指标数据 |
GetCpuloadavg15mData | read | 获取CPU十五分钟平均负载指标数据 |
GetCpuloadavg5mData | read | 获取CPU五分钟平均负载指标数据 |
GetCvmDiskUsageData | read | 获取磁盘利用率指标数据 |
GetDiskTotalData | read | 获取磁盘分区总容量指标数据 |
GetDiskUsageData | read | 获取磁盘分区已使用容量和总容量的百分比指标数据 |
GetMemUsageData | read | 获取内存利用率指标数据 |
GetMemUsedData | read | 获取内存使用量指标数据 |
InquiryPriceRunInstances | read | 创建实例询价 |
ModifyAutoScalingGroup | read | 修改伸缩组 |
ModifyDesiredCapacity | read | 修改期望实例数 |
QuickRunInstance | read | 快速创建实例(自动处理各种参数,简化创建流程) |
RebootInstances | read | 重启实例 |
ReplaceSecurityGroupPolicies | read | Replace security group rules in batch (can only replace rules in one direction per request) |
ResetInstance | read | 重装指定实例的操作系统(敏感操作,请确认后执行) |
ResetInstancesPassword | read | 重置实例密码 |
RunInstances | read | 创建实例(请注意该操作将会产生费用) |
StartInstances | read | 启动实例 |
StopInstances | read | 关闭实例 |
TerminateInstances | read | 退还实例(敏感操作,请确认后执行) |
add-issue-labels | write | 为指定Issue添加标签 |
add_timesheets | write | 填写花费工时,返回新建花费工时的数据,可以先用get_timesheets查下owner是否在spentdate已经填写过花费,如果spentdate日期已存在花费记录,则需要调用update_timesheets进行更新 |
aiPicMatting | read | 图片处理-抠图 |
aiQrcode | read | 图片处理-二维码识别-识别存储桶内二维码图片内容 |
aiSuperResolution | read | 图片处理-超分辨率 |
assessQuality | read | 图片处理-图片质量评估 |
create-group | write | 创建新组织 |
create-issue | write | 创建一个 Issue. 如需添加 Issue 标签,需要另外调用 add-issue-labels |
create-issue-comment | write | 创建一个 Issue 评论 |
create-pull | write | 创建Pull Request |
create-pull-comment | write | 创建Pull Request评论 |
create-repository | write | 创建仓库 |
createDocToPdfJob | read | 创建文档转 pdf 处理任务 |
createMediaSmartCoverJob | read | 创建媒体智能封面任务 |
create_bug | write | 创建 TAPD 缺陷 |
create_comments | write | 添加评论 |
create_issue | write | 创建新的工作项 |
create_iteration | write | 创建迭代,返回创建迭代的数据 |
create_merge_request | write | 创建合并请求。需要提供工作目录路径,会自动从指定目录的 Git 仓库获取仓库信息。如果不指定源分支,将使用当前分支;如果不指定目标分支,将使用 master。仓库路径会自动从 Git 远程仓库地址中解析。 |
create_or_update_tcases | write | 新建测试用例,返回新建测试用例的数据 |
create_project | write | 在 CODING DevOps 中创建新项目 |
create_story_or_task | write | 创建 TAPD 需求或任务 |
create_wiki | write | 新建Wiki,返回新建Wiki的数据 |
decompose_issue | read | 将一个需求拆解为多个子任务 |
delete-issue-label | destructive | 删除Issue的指定标签 |
delete-issue-labels | destructive | 删除Issue的所有标签 |
delete-workspace | destructive | 删除我的云原生开发环境 |
delete_issue | destructive | 删除工作项 |
delete_project | destructive | 删除 CODING DevOps 中的指定项目 |
deploy_folder_or_zip | write | Deploy a built frontend directory (or zip file) to EdgeOne Pages. Returns: the deployment URL and project metadata. |
deploy_html | write | Deploy HTML content to EdgeOne Pages, return the public URL |
describeDocProcessJob | read | 根据 jobid 查询指定的文档转码任务结果 |
describeMediaJob | read | 根据 jobid 查询指定的媒体智能封面任务结果 |
describe_issue | read | 查询事项详情 |
entity_relations | read | 创建需求与缺陷关联关系 |
get-component-changelog | read | 获取 TDesign 组件的变更日志,适用场景:组件库版本升级,支持多个组件同时查询 |
get-component-docs | read | 获取 TDesign 组件的文档,适用场景:代码生成和代码转换,支持多个组件同时查询 |
get-component-dom | read | 获取 TDesign 组件的 DOM 结构,适用场景:转换用户自定义的 CSS 样式, 支持多个组件同时查询 |
get-component-list | read | 获取 TDesign 所有可用的组件列表,适用场景:选择合适组件进行组合实现不存在的功能 |
get-current-repo | read | 获取当前仓库对应的CNB仓库信息 |
get-group | read | 获取指定组织信息 |
get-issue | read | 获取指定 Issue 信息 |
get-pull | read | 获取单个Pull Request详情 |
get-repository | read | 获取指定仓库信息 |
getBucket | read | 查询存储桶内的文件列表 |
getCosConfig | read | 获取COS配置, 腾讯云配置 |
getObject | read | 下载存储桶内的文件 |
getObjectUrl | read | 获取存储桶内的文件的带签名的下载链接 |
get_bug | read | 获取 TAPD 缺陷 |
get_bug_count | read | 获取 TAPD 缺陷 |
get_comments | read | 获取符合查询条件的所有评论(分页显示,默认一页30条) |
get_commit_msg | write | 获取需求/缺陷/任务的源码提交关键字,将 commit 和需求/缺陷/任务关联 |
get_entity_custom_fields | read | 获取 TAPD 需求或任务或迭代或测试用例的自定义字段配置 |
get_iterations | read | 根据项目ID(workspace_id)获取符合查询条件的所有迭代,例如可以通过名称获取迭代 id,然后作为iteration_id字段的值获取到迭代下的需求或缺陷 |
get_related_bugs | read | 获取符合查询条件的所有需求关联的缺陷ID |
get_release_info | read | 获取发布计划信息,返回符合查询条件的所有发布计划(分页显示),默认返回 30 条。可通过传 limit 参数设置,最大取 200。也可以传 page 参数翻页 |
get_stories_fields_info | read | 获取需求所有字段及候选值,返回符合查询条件的所有需求字段及候选值。 部分字段为静态候选值,建议参考下方 |
get_stories_fields_lable | read | 获取需求所有字段的中英文 |
get_stories_or_tasks | read | 获取 TAPD 需求或任务,如果没有 limit 参数,则需要同时调用 get_story_count 工具获取需求数量 |
get_story_or_task_count | read | 获取 TAPD 需求或者任务的数量 |
get_tcases | read | 获取测试用例,返回符合查询条件的所有测试用例(分页显示,默认一页30条),默认返回 30 条。可通过传 limit 参数设置,最大取 200。也可以传 page 参数翻页 |
get_timesheets | read | 获取花费工时,返回符合查询条件的所有花费工时(分页显示,默认一页30条)默认返回 30 条。可通过传 limit 参数设置,最大取 200。也可以传 page 参数翻页 |
get_todo | read | 获取用户的待办,返回给用户的时候需要说明这是待办 |
get_user_participant_projects | read | |
get_wiki | read | 获取 Wiki,返回符合查询条件的所有Wiki(分页显示,默认一页30条) |
get_workflows_all_transitions | read | 获取项目下的工作流流转细则,例如要流转到 |
get_workflows_last_steps | read | 获取工作流结束状态,一次只能获取一个项目的工作流结束状态 |
get_workflows_status_map | read | 获取工作流状态中英文名对应关系,例如想要获取所有 |
get_workitem_types | read | 获取符合查询条件的所有需求类别(分页显示,默认一页30条) |
get_workspace_info | read | 根据项目ID(workspace_id)获取项目信息,包含项目ID,项目名称,状态,创建时间,创建人等信息 |
imageInfo | read | 图片处理-获取图片信息 |
imageSearchPic | read | 根据输入的图片,从数据集中检索出与输入的图片内容相似的图片 |
imageSearchText | read | 根据输入的文本内容,从数据集中检索出与输入的文本内容相符的图片 |
job_detail | read | 获取任务详情, 参数如下: -mcpConfigFile: mcp配置文件路径,当前项目根目录查找 tca-mcp.ini(绝对路径) -jobId: 任务id |
job_list | read | 获取当前任务列表, 参数如下: -mcpConfigFile: mcp配置文件路径,当前项目根目录查找 tca-mcp.ini(绝对路径) -offset: 偏移量 -limit: 限制数量 -state: 查询任务状态, (0为等待中,1为执行中,2为关闭,3为入库中,4正在初始化, 5初始化完成 可多选,格式为1,2,3) |
list-group-repositories | read | 获取分组里当前用户有权限的仓库 |
list-groups | read | 获取当前用户拥有权限的顶层组织列表 |
list-issue-comments | read | 查询 Issue 评论列表 |
list-issue-labels | read | 查询指定Issue的标签 |
list-issues | read | 查询仓库的 Issues |
list-pull-comments | read | 列出Pull Request的评论 |
list-pulls | read | 查询仓库的Pull Requests |
list-repositories | read | 获取当前用户拥有指定权限及其以上权限的仓库 |
list-sub-groups | read | 查询当前用户在指定组织下拥有指定权限的子组织列表 |
list-workspace | read | 获取我的云原生开发环境列表 |
list_commits | read | 查询仓库分支下提交列表 |
list_depots | read | 查询某项目的仓库列表 |
list_issues | read | 查询我最近更新的工作项,可以根据其中IssueType查询,默认查询limit为20条 |
list_projects | read | 查询当前用户在 CODING DevOps 中的项目列表 |
merge-pull | write | 合并Pull Request |
putObject | read | 上传本地文件到存储桶 |
putObjectSourceUrl | read | 通过 url下载文件并将文件上传到存储桶 |
send_qiwei_message | write | 发送信息到企业微信群 |
set-issue-labels | write | 设置Issue的标签(替换所有现有标签) |
start_scan | write | 启动tca代码分析 参数如下: -mcpConfigFile: mcp配置文件路径,当前项目根目录查找 tca-mcp.ini(绝对路径) -incrScan: 是否增量扫描 -forceCreate: 如果已经存在扫描结果,是否强制启动新扫描 |
tca_issue_list | read | 获取当前项目未解决的问题列表, 参数如下: -mcpConfigFile: mcp配置文件路径,当前项目根目录查找 tca-mcp.ini(绝对路径) -state: 查询任务状态, (0为等待中,1为执行中,2为关闭,3为入库中,4正在初始化, 5初始化完成 可多选,格式为1,2,3) -severity: 问题严重程度, (0为低,1为中,2为高) -offset: 偏移量 -limit: 限制数量 |
tca_issue_report | read | tca报告链接, 参数如下: -mcpConfigFile: mcp配置文件路径,当前项目根目录查找 tca-mcp.ini(绝对路径) -jobId: 任务id, 启动分析任务后,从返回结果中获取 |
update-issue | write | 更新一个 Issue。 如需更新 Issue 标签,需要另外调用 set-issue-labels |
update-issue-comment | write | 更新一个 Issue 评论 |
update-pull | write | 更新Pull Request |
update_bug | write | 更新 TAPD 缺陷 |
update_comments | write | 更新评论,返回更新评论的数据 |
update_iteration | write | 更新迭代,返回更新迭代的数据 |
update_story_or_task | write | 更新 TAPD 需求或任务,任务的状态只有三个,open: 未开始,progressing: 进行中,done: 已完成,如果是流转需求的状态,需要调用工具先获取状态信息get_workflows_all_transitions |
update_timesheets | write | 更新花费工时,返回花费工时更新后的数据,每次只允许更新一条数据 |
update_wiki | write | 更新 TAPD wiki |
waterMarkFont | read | 生成带文字水印的图片 |
Trust audit
BLOCKgrade F · trust 50/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | FAIL |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | FAIL |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (6 observation(s))
- Network
- declared (10 observation(s))
- Shell
- declared (2 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
.env
5. Prefer the official Tencent Cloud SDK. If the SDK is unavailable, do not block on that alone; generate properly TC3-signed HTTPS requests locally instead. Read credentials only from environment var
6. Prefer the official Tencent Cloud SDK. If the SDK is unavailable, do not block on that alone; generate properly TC3-signed HTTPS requests locally instead. Read credentials only from environment var
5. Prefer the official Tencent Cloud SDK. If the SDK is unavailable, do not block on that alone; generate properly TC3-signed HTTPS requests locally instead. Read credentials only from environment var
| [ 文字识别 OCR-文档抽取(多模态版) MCP](src/ocr-wordrecognition-multimodal) | 文档智能(Document AI) 深度融合 OCR 与多模态大模型,<br>实现高精度识别、智能解析<br>与结构化信息抽取。| [Install Go Source](#通常对于-go-源码类的-mcp您可以) | [ MCP Server](src/tca-mcp-server) | Tencent Cloud Code Analysis MCP Server, based on the MCP protocol, accurately tracks and manages code quality defects, code stand
> 文档智能(Document AI) 深度融合 OCR 与多模态大模型,实现高精度识别、智能解析与结构化信息抽取。覆盖货运单证、跨境物流、快递面单、教育作业、保险理赔及国际结算等场景,助力企业自动化升级,提升运营效率与数据准确性。
# Tencent Cloud Code Analysis (TCA) MCP Server
## TCA MCP Server Usage Steps
await this.postFetch('https://otheve.beacon.qq.com/analytics/v2_upload', payload);serverURL := fmt.Sprintf("http://127.0.0.1:%s%s", ssePort, sseEndpoint)serverURL := fmt.Sprintf("http://127.0.0.1:%s%s", ssePort, sseEndpoint)mcp.Description("配置id支持:General -- 通用场景 InvoiceEng -- 国际invoice模板 WayBillEng --海运订单模板 CustomsDeclaration -- 进出口报关单 WeightNote -- 磅单 MedicalMeter -- 血压仪表识别 BillOfLading -- 海运提单 EntrustmentBook -- 海运托书delete-issue-label, delete-issue-labels, delete-workspace, delete_issue, delete_project
public.zip
.versionrc
.versionrc.json
.augment-guidelines
.opencode.json
.cnb.yml
import { version as packageVersion } from '../../package.json';import { CodingConnection } from '../../api/coding_connection.js';import { CodingDevOpsConfig } from '../../config/environment.js';import { CodingDevOpsConfig } from '../../config/environment.js';import { CodingDevOpsConfig } from '../../config/environment.js';Gates applied: no_behavioural_pass.
590795f31c64full audit observations/trust-audit/mcp-server/tencentcloudcommunity__tencent-cloud.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 590795f31c64 | BLOCK | F | 50 | first audit |
Questions
What is the Tencent Cloud MCP server?
Tencent Cloud MCP Server, Support Remote Hosted and Local Installation
What tools does Tencent Cloud expose?
148 in total: 107 read-only, 36 that write, and 5 that can delete or overwrite (delete-issue-label, delete-issue-labels, delete-workspace, delete_issue, delete_project). Every one is listed on this page with its risk.
Is Tencent Cloud safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (50/100) and found 9 critical or high issues in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 5 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Tencent Cloud need?
It reads API_TOKEN, EDGEONE_PAGES_API_TOKEN, MINIPROGRAM_PRIVATE_KEY, MINIPROGRAM_PRIVATE_KEY_PATH, TAPD_ACCESS_TOKEN, TAPD_API_PASSWORD, TCA_TOKEN, TENCENTCLOUD_SECRETID, TENCENTCLOUD_SECRETKEY, TENCENTCLOUD_SECRET_ID, TENCENTCLOUD_SECRET_KEY and TENCENTCLOUD_SESSIONTOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Tencent Cloud run?
It speaks sse, stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as tdesign-mcp-server at 0.2.0.
How current is this page?
The grade is for one exact copy of the source (590795f31c64), read on 2026-10-08. The repository is watched and re-audited when it changes.