archery-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ARCHERY_DB_NAME | Yes | The locked database name | |
| ARCHERY_BASE_URL | Yes | Archery deployment address, e.g. https://archery.example.com | |
| ARCHERY_PASSWORD | Yes | Archery password | |
| ARCHERY_USERNAME | Yes | Archery username (a dedicated least-privilege account is recommended) | |
| ARCHERY_QUERY_LIMIT | No | Default number of rows when 'limit' is not passed to query; 0 means let Archery use the maximum limit according to account permissions | 100 |
| ARCHERY_SCHEMA_NAME | No | The locked schema name; usually 'public' for PostgreSQL, can be left empty for MySQL | public |
| ARCHERY_TOTP_SECRET | No | The base32 secret for Google Authenticator when the account has two-factor authentication enabled (not the six-digit verification code) | |
| ARCHERY_INSTANCE_NAME | Yes | The locked Archery instance name (must be within the account's accessible list) |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_tablesB | 列出当前配置的 Archery 目标库中的表，可按关键字过滤表名。 Args: keyword: 可选的表名过滤关键字，大小写不敏感 |
| describe_tableB | 查看当前配置目标库中指定表的字段、类型、默认值和注释。 Args: table_name: Archery 中可访问的表名 |
| queryA | 通过 Archery 在线查询执行 SQL。 SQL 是否可执行、是否只读、是否包含敏感字段、最大返回行数、查询超时及 结果脱敏均由 Archery 的权限和策略控制。本 MCP 不修改 SQL，也不屏蔽 Archery 返回的字段。 Args: sql: 待提交给 Archery 在线查询的 SQL limit: 可选查询行数。未传时使用 ARCHERY_QUERY_LIMIT；传 0 时交给 Archery 使用其最大权限限制 |
| query_historyA | 查看当前 Archery 账号自己的在线查询历史。 Args: limit: 返回条数，必须为正整数 search: 按 SQL 内容、用户名或别名过滤的关键字 |
| sql_checkA | 检查上线 SQL 是否符合内置规范与 Archery 平台要求。 检查项（内置规范）：
同时调用 Archery 平台的 SQL 检查接口（需要账号在 Archery 的 API 用户白名单内并有上线权限）；平台检查不可用时保留本地结果并说明原因。 Args: sql: 待检查的上线 SQL，可包含多条语句（分号分隔） |
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 5 tools
Each tool targets a distinct resource/action: list_tables (enumerate), describe_table (schema detail), query (execute SQL), query_history (own past queries), and sql_check (validate deployment SQL). The only mild adjacency is query vs sql_check, but their descriptions clearly separate execution from compliance validation.
Names follow a mostly consistent snake_case pattern (list_tables, describe_table, query_history, sql_check), with the lone bare 'query' being a minor deviation from the verb_noun style. Still readable and predictable throughout.
Five tools is well-scoped for a database access server: just enough to discover schema, run queries, review history, and validate deploy SQL without redundancy. No filler tools.
Covers the core discovery-query-validate lifecycle (list, describe, query, history, sql_check). Some gaps remain, such as explicit multi-database/target selection, explain/plan inspection, or write/deploy execution, but these are largely outside a read-oriented Archery query scope.