菜单
查询 AxonHub AI Gateway 的请求记录、Trace、UsageLog，以及管理渠道、API Key、模型等资源。使用官方 @axonhub/graphql-cli 工具通过 GraphQL API 操作，支持按 traceID、model、时间、状态筛选请求，也可探索 schema、执行任意 mutation。当用户需要查看 AxonHub 请求详情、排查 LLM 调用、查看 token 消耗、追踪 Trace、或通过命令行管理渠道/Key 时使用。
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/justcyl/my-skills --skill axonhub命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
使用 Kapibala 的 OpenAI 兼容图片接口生成/编辑位图图片，默认模型 gpt-image-2。当需要生成图片、编辑已有图片、或批量生成候选图供挑选时触发。触发词："生成图片""画一张""image generation""edit this image""gpt-image-2""批量出图"。
Add, collect, design, audit, and integrate SVG elements into editable diagrams.net/draw.io scientific figures. Use when the user asks to supplement a draw.io diagram with SVG icons, mechanisms, biological objects, model components, pathway symbols, instruments, graphical abstract elements, or publication-style vector assets. Every execution must first ask the user to choose the SVG source mode: network search or self-designed SVG. For network mode, search for multiple relevant SVG candidates for later user selection, record sources and license notes, and fall back to self-designed SVG for concepts with no suitable result. For self-design mode, create restrained, editable, Nature-style SVGs from simple vector primitives. Use with .drawio, SVG, diagrams.net, 科研流程图, 论文示意图, SVG补图, 图标补充, and draw.io配图.
Two-stage workflow for publication-style scientific schematics: generate or use a raster reference with imagegen/PNG/JPG/WebP, then trace it into an editable diagrams.net/draw.io file using research-drawio-skill. Use for Nature-style scientific illustrations, graphical abstracts, model or mechanism schematics, biomedical workflows, AI architecture figures, and paper-style visual drafts that should become editable .drawio source. For complex recognizable elements, author or select dedicated SVG glyphs and compare rendered SVGs against reference crops before inserting them, instead of assembling those elements from draw.io primitives. Enforces a minimum three-round strict pixel-level export/compare/fix loop for close raster-to-draw.io tracing tasks. Use with imagegen, draw.io, diagrams.net, scientific illustration, graphical abstract, paper schematic, figure tracing, SVG glyph tracing, and drawio redraw requests.
基于 SOC 职业分类
正在显示 SKILL.md
| name | axonhub |
| description | 查询 AxonHub AI Gateway 的请求记录、Trace、UsageLog，以及管理渠道、API Key、模型等资源。使用官方 @axonhub/graphql-cli 工具通过 GraphQL API 操作，支持按 traceID、model、时间、状态筛选请求，也可探索 schema、执行任意 mutation。当用户需要查看 AxonHub 请求详情、排查 LLM 调用、查看 token 消耗、追踪 Trace、或通过命令行管理渠道/Key 时使用。 |
通过官方 @axonhub/graphql-cli 工具操作 AxonHub GraphQL API。无需安装，直接 npx 使用。
官方 skill 来源：
looplj/axonhub仓库skills/axonhub-cli/SKILL.md（PR #1149，v0.9.x+）
curl 和 jqhttp://localhost:8090）# 验证工具可用
npx @axonhub/graphql-cli --version
export AXONHUB_EMAIL="your-email@example.com"
export AXONHUB_PASSWORD="your-password"
AXONHUB_URL="http://localhost:8090"
AXONHUB_TOKEN=$(curl -s -X POST "${AXONHUB_URL}/admin/auth/signin" \
-H "Content-Type: application/json" \
-d "{\"email\":\"${AXONHUB_EMAIL}\",\"password\":\"${AXONHUB_PASSWORD}\"}" \
| jq -r '.token')
# 验证
[ "$AXONHUB_TOKEN" != "null" ] && [ -n "$AXONHUB_TOKEN" ] \
&& echo "✅ 登录成功" || echo "❌ 登录失败"
Token 有效期 7 天，过期后重新执行上面的命令即可。
只需配置一次，之后所有查询用 -e axonhub 引用：
npx @axonhub/graphql-cli add axonhub \
--url "${AXONHUB_URL}/admin/graphql" \
--description "AxonHub GraphQL API"
npx @axonhub/graphql-cli login axonhub \
--type token \
--token "${AXONHUB_TOKEN}"
# 验证
npx @axonhub/graphql-cli list --detail
不确定字段名时，先用 find 探索再查询，避免硬猜字段：
# 列出所有可用 query
npx @axonhub/graphql-cli find -e axonhub --query
# 列出所有可用 mutation
npx @axonhub/graphql-cli find -e axonhub --mutation
# 按关键词搜索（只显示名字）
npx @axonhub/graphql-cli find request -e axonhub
npx @axonhub/graphql-cli find channel -e axonhub
# 加 --detail 看完整字段定义
npx @axonhub/graphql-cli find request -e axonhub --detail
# 查 mutation 所需 input 类型
npx @axonhub/graphql-cli find CreateChannel -e axonhub --input --detail
规则：先不加
--detail看名字概览，确认是目标后再加--detail看完整定义，避免输出过多。
npx @axonhub/graphql-cli query \
'{ requests(first: 20, orderBy: {field: CREATED_AT, direction: DESC}) {
totalCount
edges { node { id modelID status stream createdAt metricsLatencyMs metricsFirstTokenLatencyMs } }
} }' \
-e axonhub | jq '.requests'
TRACE_ID="at-demo-123"
npx @axonhub/graphql-cli query \
"{ requests(where: {traceID: \"$TRACE_ID\"}, orderBy: {field: CREATED_AT, direction: ASC}) {
edges { node { id modelID status createdAt metricsLatencyMs } }
} }" \
-e axonhub | jq '.requests.edges[].node'
REQUEST_ID="12345"
npx @axonhub/graphql-cli query \
"{ requests(where: {id: \"$REQUEST_ID\"}) {
edges { node { id modelID status requestBody responseBody metricsLatencyMs } }
} }" \
-e axonhub | jq '.requests.edges[0].node'
若
requestBody/responseBody为 null（内容已外存），改用 REST：curl -s "${AXONHUB_URL}/admin/requests/${REQUEST_ID}/content" \ -H "Authorization: Bearer ${AXONHUB_TOKEN}"
npx @axonhub/graphql-cli query \
'{ requests(first: 20, where: {modelIDContainsFold: "claude"},
orderBy: {field: CREATED_AT, direction: DESC}) {
totalCount
edges { node { id modelID status createdAt metricsLatencyMs } }
} }' \
-e axonhub | jq '.requests'
npx @axonhub/graphql-cli query \
'{ requests(first: 20, where: {status: failed},
orderBy: {field: CREATED_AT, direction: DESC}) {
totalCount
edges { node { id modelID status createdAt } }
} }' \
-e axonhub | jq '.requests'
状态枚举：pending | processing | completed | failed | canceled
npx @axonhub/graphql-cli query \
'{ requests(first: 50,
where: {createdAtGTE: "2025-04-25T00:00:00Z", createdAtLT: "2025-04-26T00:00:00Z"},
orderBy: {field: CREATED_AT, direction: DESC}) {
totalCount
edges { node { id modelID status createdAt metricsLatencyMs } }
} }' \
-e axonhub | jq '.requests'
npx @axonhub/graphql-cli query \
'{ requests(first: 20,
where: { and: [
{ status: failed },
{ modelIDContainsFold: "gpt" },
{ createdAtGTE: "2025-04-26T10:00:00Z" }
]},
orderBy: {field: CREATED_AT, direction: DESC}) {
totalCount
edges { node { id modelID status createdAt } }
} }' \
-e axonhub | jq '.requests'
npx @axonhub/graphql-cli query \
'{ usageLogs(first: 20, orderBy: {field: CREATED_AT, direction: DESC}) {
edges { node { id modelID promptTokens completionTokens totalTokens totalCost createdAt } }
} }' \
-e axonhub | jq '.usageLogs.edges[].node'
npx @axonhub/graphql-cli query \
'{ traces(first: 10, orderBy: {field: CREATED_AT, direction: DESC}) {
edges { node { id traceID createdAt
requests { edges { node { id modelID status metricsLatencyMs } } }
} }
} }' \
-e axonhub | jq '.traces.edges[].node'
npx @axonhub/graphql-cli query \
'{ systemStatus { isInitialized }
systemVersion { version commit uptime } }' \
-e axonhub | jq '.'
npx @axonhub/graphql-cli query \
'{ queryChannels(input: { first: 20 }) {
edges { node { id name type status supportedModels } }
} }' \
-e axonhub | jq '.queryChannels.edges[].node'
CHANNEL_ID="1"
npx @axonhub/graphql-cli mutate \
"mutation { testChannel(input: { channelID: \"$CHANNEL_ID\" }) {
success latency message error
} }" \
-e axonhub | jq '.testChannel'
npx @axonhub/graphql-cli mutate \
'mutation { updateChannelStatus(id: "1", status: enabled) { id name status } }' \
-e axonhub
npx @axonhub/graphql-cli query \
'{ me { id email firstName lastName isOwner scopes } }' \
-e axonhub | jq '.me'
# 第一页，同时取 pageInfo
npx @axonhub/graphql-cli query \
'{ requests(first: 20, orderBy: {field: CREATED_AT, direction: DESC}) {
pageInfo { hasNextPage endCursor }
totalCount
edges { node { id modelID createdAt } }
} }' \
-e axonhub | jq '.'
# 翻下一页（把 endCursor 填入 after）
npx @axonhub/graphql-cli query \
'{ requests(first: 20, after: "<endCursor>",
orderBy: {field: CREATED_AT, direction: DESC}) {
pageInfo { hasNextPage endCursor }
edges { node { id modelID createdAt } }
} }' \
-e axonhub | jq '.'
| 字段 | 说明 |
|---|---|
id | 请求 ID |
modelID | 模型名 |
status | pending/processing/completed/failed/canceled |
stream | 是否流式 |
source | api/playground/test |
requestBody | 请求体（可能 null，已外存时用 REST 下载） |
responseBody | 响应体（同上） |
metricsLatencyMs | 总延迟（ms） |
metricsFirstTokenLatencyMs | 首 token 延迟（ms） |
metricsReasoningDurationMs | 推理时长（ms） |
traceID | 所属 Trace（外键 ID） |
channelID | 路由到的渠道 ID |
createdAt | 请求时间 |