byerisk-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BYERISK_API_KEY | Yes | API Key created in the ByeRisk console, e.g., brsk_live_... | |
| BYERISK_BASE_URL | No | Base URL for the ByeRisk API. Default is https://www.byerisk.com/api | https://www.byerisk.com/api |
| BYERISK_END_TENANT | No | Sub-tenant ID for data isolation. 1-64 characters consisting of letters, digits, '_', '.', ':', '-'. |
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 |
|---|---|
| check_textA | 对一段文案做发布前合规预检，返回「必须修改 / 影响推流」两档风险明细、法条依据与修改建议。适用于短视频脚本、口播稿、短剧剧本、投放素材文案、商品标题详情等任何要发到内容平台的中文文案。每次调用都会扣除账户积分，同一段文案不要重复提交。 |
| fix_textA | 对一次已完成的文案检测触发 AI 改写，产出规避了命中风险的合规版本。需要先用 check_text 拿到检测 id。会扣除修复积分。 |
| check_imageA | 检测图片是否含广告违规、违规文字（OCR）、涉政、暴力、色情、侵权 Logo 等风险。支持传本机文件路径（会自动上传，用户的图通常就在本机）或公网直链。短剧产品线没有图片检测。会扣积分。 |
| check_videoA | 检测视频的画面、音频与口播话术合规风险，返回带时间点的风险明细。支持传本机文件路径（会自动上传）或公网直链。按视频时长计费（1 积分/秒），长视频要等数分钟。若该视频的脚本此前用 check_text 检测过，把那次的 id 传进 textCheckId 可额外做口播漂移检测（比对实际口播与过审脚本的偏离片段）。 |
| get_check_resultA | 按 id 取一次已提交检测的结果。用于上一次调用等待超时、或需要回看历史检测的场景。不扣积分。 |
| get_accountA | 查询 ByeRisk 账户的积分余额、订阅状态与本月用量。检测报 402 积分不足时可以用它确认余额。不扣积分。 |
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 6 tools
Each tool targets a distinct function: account status, text check, text fix, image check, video check, and result retrieval. There is no overlap between check_text and check_image/video as they handle different media types, and fix_text clearly depends on check_text output.
All tool names follow a consistent verb_noun pattern in snake_case: get_account, check_text, fix_text, check_image, check_video, get_check_result. The verbs are meaningful (get, check, fix) and the pattern is uniform across the entire set.
With 6 tools, the server is well-scoped for a compliance-checking service. Each tool covers a distinct capability (account, text, image, video, fixing, and result lookup) without unnecessary bloat or missing essentials.
The tool surface provides complete lifecycle coverage for the domain: submission of checks for all major content types (text, image, video), a fix operation for text, asynchronous result retrieval, and account monitoring. There are no dead ends; every operation leads to a result accessible via get_check_result.