Volcengine RDS PostgreSQL
v1.0.0使用火山引擎 RDS PostgreSQL,帮助用户完成 RDS PostgreSQL 相关的实例管理、数据库操作、账号管理和运维任务,可直接调用 uv run ./scripts/call_rds_postgresql.py 脚本获取实时结果。
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description, SKILL.md, requirements.txt, and the Python script all focus on Volcengine RDS PostgreSQL and VPC operations. Requesting VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY is appropriate and expected for calling the Volcengine SDK APIs invoked in the script.
Instruction Scope
SKILL.md directs the agent to run the included script (via `uv run` or `python`) and to set the Volcengine AK/SK and optional region. The script only constructs SDK requests for RDS and VPC endpoints and prints results to stdout/stderr. There are no instructions to read unrelated files, other environment variables, or to transmit data to unexpected external endpoints.
Install Mechanism
This is instruction-only (no platform install spec). The included requirements.txt and the script reference the official-looking `volcengine-python-sdk[rdspostgresqlv2,vpc]` package; dependencies must be installed by the user (or by `uv` execution). This is low-risk, but users should ensure they trust the PyPI package and have `uv` available.
Credentials
Only VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY are required (and an optional region). Those credentials directly map to the documented API calls. No unrelated secrets or broad system credentials are requested.
Persistence & Privilege
The skill is not marked always:true and does not request persistent system-wide privileges or modify other skills' configs. The script runs on-demand and does not install background services.
Assessment
This skill appears coherent for managing Volcengine RDS PostgreSQL. Before enabling: (1) provide only a least-privilege AK/SK — ideally read-only permissions for the RDS and VPC actions the skill needs, (2) review the volcengine-python-sdk package on PyPI (and its maintainers) before installing dependencies, (3) ensure you have the `uv` runner or install Python deps in an isolated environment, and (4) avoid using production credentials with write/delete permissions unless you explicitly need full management operations. The script itself only calls Volcengine SDK APIs and prints results; if you require higher assurance, inspect the full script and test with a non-production account first.Like a lobster shell, security has layers — review code before you run it.
Runtime requirements
🐘 Clawdis
OSmacOS · Linux
Binsuv
EnvVOLCENGINE_ACCESS_KEY, VOLCENGINE_SECRET_KEY
latest
Skill 概览
本 Skill 用于在对话中充当 火山引擎 RDS PostgreSQL 的智能运维代理:
- 理解用户的自然语言需求(中文或英文),识别是否与 RDS PostgreSQL 相关;
- 直接调用内置脚本
scripts/call_rds_postgresql.py实时查询 RDS PostgreSQL 并获取结果; - 当获取到结果或用户粘贴错误信息时,进一步解释、诊断并给出后续建议。
工作模式:
- 使用
scripts/call_rds_postgresql.py脚本直接获取 RDS PostgreSQL 的实时响应
运行方式: 脚本支持两种运行方式:
# 方式 1: 使用 uv (推荐,自动管理依赖)
uv run ./scripts/call_rds_postgresql.py [action] [options]
# 方式 2: 使用 python (需要预先安装依赖)
python ./scripts/call_rds_postgresql.py [action] [options]
标准使用流程
-
确认任务类型与参数
- 判断用户意图:查询实例列表、查看实例详情、管理数据库、管理账号、查看参数配置、创建实例等。
- 收集必要参数(如未指定则使用默认值):
--region:地域 ID(默认cn-beijing)--action:操作类型(如list-instances、describe-instance、list-databases等)--instance-id:实例 ID(部分操作必需)
-
构造查询并调用脚本
- 示例(以下命令可使用
uv run或python运行):# 查询实例列表 uv run ./scripts/call_rds_postgresql.py list-instances # 或 python ./scripts/call_rds_postgresql.py list-instances # 查询指定实例详情 uv run ./scripts/call_rds_postgresql.py describe-instance --instance-id postgres-xxx # 或 python ./scripts/call_rds_postgresql.py describe-instance --instance-id postgres-xxx # 查询实例的数据库列表 uv run ./scripts/call_rds_postgresql.py list-databases --instance-id postgres-xxx # 或 python ./scripts/call_rds_postgresql.py list-databases --instance-id postgres-xxx # 查询实例的账号列表 uv run ./scripts/call_rds_postgresql.py list-accounts --instance-id postgres-xxx # 或 python ./scripts/call_rds_postgresql.py list-accounts --instance-id postgres-xxx # 查询实例参数 uv run ./scripts/call_rds_postgresql.py list-parameters --instance-id postgres-xxx # 或 python ./scripts/call_rds_postgresql.py list-parameters --instance-id postgres-xxx # 查询 VPC 列表(用于创建实例) uv run ./scripts/call_rds_postgresql.py list-vpcs # 或 python ./scripts/call_rds_postgresql.py list-vpcs # 查询子网列表 uv run ./scripts/call_rds_postgresql.py list-subnets --vpc-id vpc-xxx --zone-id cn-beijing-a # 或 python ./scripts/call_rds_postgresql.py list-subnets --vpc-id vpc-xxx --zone-id cn-beijing-a
- 示例(以下命令可使用
-
解析结果并后续处理
- 将 RDS PostgreSQL 的响应用自然语言解释给用户;
- 如返回包含敏感操作,评估风险并提醒:
- 避免在生产环境直接执行高风险操作(如删除实例、删除数据库等);
- 建议在测试环境验证或做好备份。
工具脚本使用说明
支持的操作(Actions)
| 操作 | 说明 | 必需参数 |
|---|---|---|
list-instances | 查询 RDS PostgreSQL 实例列表 | 无 |
describe-instance | 查询指定实例详情 | --instance-id |
list-databases | 查询实例的数据库列表 | --instance-id |
list-accounts | 查询实例的账号列表 | --instance-id |
list-parameters | 查询实例的参数配置 | --instance-id |
list-parameter-templates | 查询参数模板列表 | 无 |
describe-parameter-template | 查询参数模板详情 | --template-id |
list-vpcs | 查询 VPC 列表 | 无 |
list-subnets | 查询子网列表 | --vpc-id |
get-price | 查询实例价格 | 多个配置参数 |
命令行参数
| 参数 | 说明 | 默认值 |
|---|---|---|
action | 操作类型(必需) | - |
--region / -r | 火山引擎地域 ID | cn-beijing |
--instance-id / -i | 实例 ID | 无 |
--page-number | 分页页码 | 1 |
--page-size | 每页记录数 | 10 |
--output / -o | 输出格式(json/table) | json |
注意:
--region、--page-number、--page-size等全局参数必须放在子命令之前,例如:uv run ./scripts/call_rds_postgresql.py --region cn-shanghai list-instances放在子命令之后会被忽略(argparse 解析限制)。
输出格式
脚本会将查询信息输出到 stderr,将结果输出到 stdout,便于分离日志和结果:
[操作] list-instances
[地域] cn-beijing
============================================================
[查询结果]
<实际结果内容>
常见使用场景
1. 查看所有实例
uv run ./scripts/call_rds_postgresql.py list-instances
2. 查看实例详情
uv run ./scripts/call_rds_postgresql.py describe-instance --instance-id postgres-xxx
3. 查看实例的数据库
uv run ./scripts/call_rds_postgresql.py list-databases --instance-id postgres-xxx
4. 查看实例的账号
uv run ./scripts/call_rds_postgresql.py list-accounts --instance-id postgres-xxx
5. 查看实例参数配置
uv run ./scripts/call_rds_postgresql.py list-parameters --instance-id postgres-xxx
6. 按版本过滤实例
uv run ./scripts/call_rds_postgresql.py list-instances --db-engine-version PostgreSQL_14
7. 创建实例前查询网络信息
# 先查询 VPC
uv run ./scripts/call_rds_postgresql.py list-vpcs
# 再查询子网
uv run ./scripts/call_rds_postgresql.py list-subnets --vpc-id vpc-xxx --zone-id cn-beijing-a
环境变量配置
-
获取火山引擎访问凭证:参考 用户指南 获取 AK/SK
-
配置以下环境变量:
export VOLCENGINE_ACCESS_KEY="your-access-key"
export VOLCENGINE_SECRET_KEY="your-secret-key"
export VOLCENGINE_REGION="cn-beijing" # 可选,默认 cn-beijing
权限说明:
list-vpcs和list-subnets调用的是 VPC 服务接口,AK/SK 需要同时授予 RDS PostgreSQL 和 VPC(vpc:DescribeVpcs、vpc:DescribeSubnets)两个服务的权限, 否则会返回AccessDenied 403错误。
Comments
Loading comments...
