From 33d79194b81d6ac375da794f2594ff4b5e86cf34 Mon Sep 17 00:00:00 2001 From: api-gateway Date: Wed, 30 Sep 2026 15:26:12 +0800 Subject: [PATCH] =?UTF-8?q?feat(ui):=20=E7=95=8C=E9=9D=A2=E6=8F=90?= =?UTF-8?q?=E7=A4=BA=E7=99=BD=E8=AF=9D=E5=8C=96=20+=20=E6=B8=85=E7=90=86?= =?UTF-8?q?=E6=9C=AA=E5=BC=95=E7=94=A8=E7=9A=84=E6=AD=BB=E9=94=AE=EF=BC=88?= =?UTF-8?q?=E7=89=88=E6=9C=AC=202.7.0=20=E2=86=92=202.7.1=EF=BC=89?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 老板要求「检查所有界面提示,客户不需要知道的不用提示」, 二次确认口径:普通成员看不到内部实现;术语要换成大白话 (「管理员也不懂熔断器是啥」)。 - 熔断术语白话化:熔断器/熔断状态 → 「供应商自动保护」; CLOSED/HALF_OPEN/OPEN → 「正常 / 恢复中 / 已暂停」; 探测中 → 「检测中」;熔断冷却 → 「暂停时长」;窗口/阈值 → 「统计次数/触发次数」 (指标页、仪表盘、API 文档端点说明) - 智能体页去内部实现暴露: · 删掉 cfgWorkspace 整行(原显示内部 agent_id) · cfgEngine 值由「qwenpaw · https://...」改为文案「由平台提供」 · 引擎就绪/不可达 → 「可用/暂不可用」;引擎连接中断 → 「连接中断」 · slug → 「标识」;“显示名与 slug 可改” → “显示名与标识可改” · 去掉"在服务器上"等部署细节措辞 - 其它白话化:in-flight → “当前并发”;「开始与网关对话」→「开始对话」; “参数由网关固定/上游硬约束” → “参数已固定”; “上游一次只出 1 张/上游没有返回图片” → 去"上游"措辞; “下载走网关代拉” → “下载更省心”;“协议线” → “格式” - 保留(老板确认):API 文档页的技术契约(Bearer/400/路径格式名)、 配置页的文件名与变量名(gateway.yaml/.env/token_env) - 删除 120 个未被引用的死键(gc 段早期表格化编辑设计残留: f_*/h_*/circuit/tab*/pKey 等,页面只用 YAML 编辑器与 env 两块) 及 subHint(暴露引擎名+内部 URL 的孤儿键) i18n 869 → 749 键(中英对称)。ci_check.sh 六道门禁全绿。 --- app/main.py | 3 +- frontend/src/locales/en.ts | 200 ++++++------------------------ frontend/src/locales/zh.ts | 226 ++++++++-------------------------- frontend/src/views/Agents.vue | 4 +- 4 files changed, 92 insertions(+), 341 deletions(-) diff --git a/app/main.py b/app/main.py index ea76e8a..d55830c 100644 --- a/app/main.py +++ b/app/main.py @@ -338,8 +338,9 @@ _docs_on = config.server.docs_enabled # 安全默认关(report: /docs、/ope # 三处,/health 干脆没返回版本 —— 前端「网关版本」一直显示 v-。 # 2026-09-29: 2.4.0 → 2.4.1(前端间距规范统一);2.4.1 → 2.5.0(新增「智能体」功能模块);2.5.0 → 2.5.1(智能体支持图片附件);2.5.1 → 2.5.2(智能体页间距与对齐优化);2.5.2 → 2.5.3(修 KPI 与下方块之间 0 间距);2.6.0 → 2.6.1(模型切换挪到对话框左下角)。 # 2026-09-30: 2.6.1 → 2.7.0(按老板要求整个删除「火山方舟 Coding Plan」供应商:配置/模型/代码/文档/前端下拉/环境变量全清)。 +# 2026-09-30: 2.7.0 → 2.7.1(界面提示白话化:熔断器→供应商自动保护、引擎→运行环境、slug→标识;删 120 个未引用死键)。 # ★ 版本号纪律:每次改动都要往上加,小改动加小版本号、大改动加大版本号(README 有专章)。 -VERSION = "2.7.0" +VERSION = "2.7.1" app = FastAPI( title="API Gateway", diff --git a/frontend/src/locales/en.ts b/frontend/src/locales/en.ts index c44ad89..0577e8a 100644 --- a/frontend/src/locales/en.ts +++ b/frontend/src/locales/en.ts @@ -21,10 +21,10 @@ export default { more: 'More', subdashboard: 'Gateway overview · refreshed every 8s', subchat: 'Vision-ready · multi-provider direct routing', - subimages: 'Params on the left, history on the right · downloads proxied by the gateway', + subimages: 'Params on the left, history on the right · easy downloads', subvideos: '720P · 4–12s · async polling', subagents: 'Assistants that get work done', - submodels: 'Providers / models / protocol lines', + submodels: 'Providers / models / formats', submetrics: 'Usage & cost · isolated per account', subkeys: 'Up to 3 per account · stored in plaintext, copyable anytime', subusers: 'Accounts & roles · admin can inspect others’ logs and assets', @@ -93,8 +93,8 @@ export default { provShare: 'Provider success share', latency: 'Model latency P50 / P95 (s)', providerStatus: 'Provider status', - mediaCircuit: 'Media circuit', - circuit: 'Circuit', + mediaCircuit: 'Media service', + circuit: 'Auto-protection', modelCountShort: 'Models', keySet: 'Key set', name: 'Name', @@ -105,10 +105,10 @@ export default { gwVersion: 'Gateway version', concurrency: 'In-flight {cur} / peak {peak}', latTitle: 'Model latency P50 / P95 (s)', - circ_closed: 'CLOSED', - circ_half_open: 'HALF_OPEN', - circ_open: 'OPEN', - circ_not_tracked: 'Not tracked', + circ_closed: 'Normal', + circ_half_open: 'Recovering', + circ_open: 'Paused', + circ_not_tracked: 'Not monitored', noData: 'No data', requests: 'Total requests', successRate: 'Success rate', @@ -125,7 +125,7 @@ export default { noModels: 'No calls recorded yet', share: 'Share', live: 'Live', - byProtocol: 'Protocol lines', + byProtocol: 'Format coverage', byProtocolSub: 'Models per line', modelCountN: '{n} models', }, @@ -142,17 +142,17 @@ export default { fail: 'Fail', ok: 'ok', none: 'None', - circuitStatus: 'Circuit breaker status', + circuitStatus: 'Provider auto-protection', provider: 'Provider', state: 'State', recentFailures: 'Recent failures', window: 'Window', threshold: 'Threshold', - cooldown: 'Cooldown(s)', - probing: 'Probing', - stateClosed: 'CLOSED', - stateHalf: 'HALF_OPEN', - stateOpen: 'OPEN', + cooldown: 'Pause(s)', + probing: 'Checking', + stateClosed: 'Normal', + stateHalf: 'Recovering', + stateOpen: 'Paused', loadErr: 'Failed to load metrics', requests: 'Total requests', failCount: 'Failures', @@ -165,7 +165,7 @@ export default { inflightPeak: 'In flight / peak', failRateNote: 'failure rate {v}%', snapshot: 'Live snapshot', - circuitCard: 'Circuit breakers', + circuitCard: 'Auto-paused after repeated failures, auto-recovers shortly', byProvider: 'Calls by provider (success / fail)', }, logs: { @@ -238,7 +238,7 @@ export default { params: 'Parameters', clear: 'Clear', generating: 'Generating…', - empty: 'Start a conversation with the gateway (pick a model below)', + empty: 'Start a conversation (pick a model below)', servedBy: 'by {provider}', requestedTo: 'requested {req} → served {served}', reasoning: 'Reasoning', @@ -253,7 +253,7 @@ export default { imgMax: 'At most {n} images per message', imgPickErr: 'Image processing failed: {msg}', visionWarn: 'Model {model} does not accept image input — images will be ignored or rejected. Switch to a vision-capable model.', - paramLocked: 'Parameters pinned by the gateway: {p} (upstream hard limit; what you set here is overridden before sending)', + paramLocked: 'Parameters pinned: {p} (what you set here is overridden before sending)', useVision: 'Switch to {model}', waiting: 'Awaiting input…', send: 'Send', @@ -282,11 +282,11 @@ export default { refPickErr: 'Image processing failed: {msg}', refsPH: 'Image-to-image / multi-image: image URLs or data:, one per line (empty = text-to-image)', count: 'Count', - countTip: 'The upstream API returns 1 image per call; multiple images are generated one by one', + countTip: 'Only 1 image per call; multiple images are generated one by one', genProgress: 'Generating {i}/{n}', partialFail: 'Generated {done}, image #{i} failed: {msg}', refsUsed: '{n} reference image(s)', - emptyResp: 'The upstream returned no image', + emptyResp: 'No image returned', placeholder: 'Describe the image to generate (style / subject / composition)', servedBy: 'Generated by {provider}', clear: 'Clear history', @@ -317,7 +317,7 @@ export default { uploadAudio: 'Upload audio', useUrl: 'Use URL', urlPH: 'or paste an image URL', - uploadTip: 'Uploaded material is compressed to a data URI in the browser and sent directly (no public URL needed)', + uploadTip: 'Uploaded material is compressed in the browser and sent directly (no public URL needed)', pickErr: 'Failed to process material: {msg}', refMax: 'At most {n} reference images', audMax: 'At most {n} reference audios', @@ -371,7 +371,6 @@ export default { agents: { title: 'Agents', sub: 'Assistants that get work done — each has its own params and sessions', - subHint: '{n} agents · engine {engine} ({url}) · {scope}', scopeAdmin: 'Admins only', scopeOpen: 'Available to {n}', new: 'New', @@ -389,25 +388,25 @@ export default { statusRetired: 'Archived', newChat: 'New chat', thisChat: 'This chat', - engineReady: 'Engine ready', - engineDown: 'Engine unreachable', - engineDownHint: 'Engine not ready, please try again later', + engineReady: 'Available', + engineDown: 'Unavailable', + engineDownHint: 'Not reachable right now, please try again later', retry: 'Retry', stop: 'Stop', stopRun: 'Background', running: 'Generating', - inputPlaceholder: 'Say something… this agent can read/write files and run commands on the server', + inputPlaceholder: 'Say something… this agent can read/write files and run commands', execSteps: 'Ran {n} steps', execRunning: 'Running', tokens: 'Tokens', elapsed: 'Elapsed', toolsUsed: 'Tools', - interrupted: '⚠ Engine connection lost, this turn is incomplete', + interrupted: '⚠ Connection lost, this turn is incomplete', cfgBasic: 'Basics', - cfgBasicSub: 'Name and slug are editable', + cfgBasicSub: 'Name and ID are editable', cfgName: 'Name', - cfgSlug: 'slug', - cfgSlugHint: 'Must not clash with an existing model id or alias', + cfgSlug: 'ID', + cfgSlugHint: 'Must not clash with an existing model name', cfgDesc: 'Description', cfgModel: 'Model', cfgModelPick: 'Pick a model', @@ -415,8 +414,9 @@ export default { cfgModelSwitched: 'Switched to {model}', cfgModelAtChat: 'Switch it at the bottom-left of the chat box (shared by all agents, takes effect immediately)', cfgExec: 'Execution', - cfgExecSub: 'Affects server safety, read-only for now', - cfgEngine: 'Engine', + cfgExecSub: 'Touches runtime safety, read-only for now', + cfgEngine: 'Runtime', + cfgEngineVal: 'Provided by the platform', cfgWorkspace: 'Workspace', cfgParams: 'Parameters', cfgTemp: 'temperature', @@ -440,7 +440,7 @@ export default { limitReached: 'Limit reached ({n})', limitHint: 'Limit reached ({n}); archive unused agents first', sharedNote: 'Each agent has its own params and sessions; the model is shared', - riskNote: 'This agent can read/write files and run commands on the server', + riskNote: 'This agent can read/write files and run commands', runsTotal: 'Calls', runsTools: 'Tool calls', runsTokens: 'tokens', @@ -469,7 +469,7 @@ export default { aiLabel: 'Assistant', runsEmpty: 'No runs yet', formName: 'Name', - formSlug: 'slug', + formSlug: 'ID', formDesc: 'Description', formTemp: 'temperature (optional)', formSuffix: 'Role suffix (optional)', @@ -502,7 +502,7 @@ export default { freeLimitedTag: 'Limited-time free', freeLimitedHint: 'Officially "limited-time free": the offer has an expiry — the daily sync updates this label when the vendor changes it', paramTag: 'Params locked', - paramLocked: 'Parameters pinned by the gateway: {p} (upstream hard limit; values from clients and the speed test are overridden)', + paramLocked: 'Parameters pinned: {p} (values from clients and the speed test are overridden)', allTextModels: 'All text models', providerCard: 'Providers', keyOk: 'Key configured', @@ -517,21 +517,21 @@ export default { failed: 'Failed', speedtestDone: 'Speed test done, {n} models', speedtestFail: 'Speed test failed', - speedtestHint: 'Latency probed per protocol line', + speedtestHint: 'Latency probed per format', textModelCount: '{n} text models', // Round 27: the supported-formats column (three protocol lines). Line names are the // vendors' own endpoint names, so they are not translated. protocols: 'Supported formats', - protocolHint: 'Which lines this model accepts: chat/completions (OpenAI-compatible), messages (Anthropic), responses (OpenAI Responses). Any other format is rejected with 400 at the gateway edge — the gateway never translates between formats and never silently swaps in another model', + protocolHint: 'Which call formats this model accepts: chat/completions (OpenAI-compatible), messages (Anthropic), responses (OpenAI Responses). Using any other format returns 400 — requests are never auto-translated and another model is never silently swapped in', protocolNa: 'N/A', - protocolNaHint: 'Image / video models use the dedicated /v1/images/generations and /v1/videos endpoints and are not tied to the three protocol lines', + protocolNaHint: 'Image / video models use the dedicated /v1/images/generations and /v1/videos addresses and don\'t use the three formats above', protocolHintRow: 'Supported formats: {lines} ', probedLine: 'probed via {p}', // Round 31: virtual model id (alias) — clients always send model: wjd, the gateway // forwards to the model an admin picks. An alias is not a model: it never shows up in // the speed-test list or the model counts, only in this card. aliasCard: 'Virtual model id (alias)', - aliasLine: 'Only the chat/completions (OpenAI) format is supported: calling {id} on /v1/messages or /v1/responses is rejected with 400 at the gateway (it never translates formats).', + aliasLine: 'Only the chat/completions (OpenAI) format is supported: calling {id} on /v1/messages or /v1/responses returns 400 (formats are never auto-translated).', aliasInherit: 'Vision, forced params and pricing all follow the target model — the alias stores no configuration of its own, so switching the target switches everything.', aliasName: 'Alias id', aliasTarget: 'Target model', @@ -664,7 +664,7 @@ export default { epDownload: 'Download generated images / videos through the gateway (bypasses upstream hotlink protection and expiry; allow-listed hosts only)', epHistory: 'Read your own history (images / videos / chats)', epLogs: 'Your own call log', - epMetrics: 'Metrics snapshot (incl. circuit breaker states)', + epMetrics: 'Metrics snapshot (incl. provider protection state)', epHealth: 'Health check (version / provider status)', exChat: 'Example · chat', exMessages: 'Example · chat (Anthropic Messages format)', @@ -768,21 +768,11 @@ export default { otherKeys: 'Everything else', otherKeysTip: 'Read per request — effective immediately after saving (no restart, no reload wait)', saveTip: 'Saving rewrites the whole config file from this page (the old file is backed up as .bak first); secrets are never sent to the browser — only whether they are set.', - viewYaml: 'View config file', save: 'Save & apply', - saved: 'Saved and applied', - discard: 'Discard changes', editorTitle: 'Config file (gateway.yaml)', editorSub: 'Edit the full YAML; click Save to apply', - exportCfg: 'Export config', - importCfg: 'Import config', - importCfgOk: 'Config imported into editor — click Save to apply', envTitle: 'API key env vars', envSub: 'Stored in data/.env; hidden by default, show / copy / edit', - envExport: 'Export vars', - envImport: 'Import vars', - envImportErr: 'Import vars failed', - envEmpty: 'No provider uses token_env', envSet: 'Set', envMissing: 'Missing', envShow: 'Show', @@ -803,119 +793,9 @@ export default { restoreInvalid: 'Invalid backup file', errTitle: 'Save failed — nothing was changed', - needRestart: 'restart', - restartApplied: 'Saved. These need an api-gateway restart to take effect: {keys}', - savedRestart: 'Saved and applied', - addItem: 'Add', - addModel: 'Add model', - addProvider: 'Add provider', - addPrice: 'Add pricing', - delete: 'Delete', - delProvider: 'Delete provider', - delConfirm: 'Delete "{name}"? It takes effect once you save.', - needKey: 'Provider key cannot be empty', - dupKey: '"{name}" is duplicated — please rename', - newProvider: '(new provider)', - unlimited: 'empty = unlimited', - useServerTimeout: 'empty = server.timeout', - mediaEmpty: 'empty = unsupported', - modelId: 'model id', - modelDisplay: 'display name', - hdrKey: 'header', - hdrVal: 'value', - tokenSet: 'key configured', - tokenEnvMissing: 'env var not set', - tokenKeep: 'empty = keep current', - tokenEmpty: 'empty = none', - jump: 'Jump to', - tabServer: 'Server & security', - tabServerSub: 'Binding, timeout/retry, rate limits, CORS, session TTL and other runtime switches', - tabProviders: 'Providers & models', - tabProvidersSub: 'Key source, base URL, image/video endpoints and the model list per provider', - tabLog: 'Logs & pricing', - tabLogSub: 'Log level/format/capacity plus model pricing (CNY per million tokens)', - tabDownload: 'Download proxy', - tabDownloadSub: 'CDN hosts allowed for the download proxy and the per-file size limit', - circuit: 'Provider circuit breaker', - h_circuit: 'A provider failing to the threshold is cooled down; its requests get 503', - f_c_enabled: 'Enable circuit breaker', - f_c_failure_threshold: 'Failure threshold', - f_c_cooldown: 'Cooldown (s)', - f_c_window: 'Window (s)', - h_c_enabled: 'Off = never skip a failing provider', - h_c_failure_threshold: 'Opens the circuit once failures reach this count in the window', - h_c_cooldown: 'While open the provider is skipped; it half-opens automatically after the cooldown', - h_c_window: 'Counts failures inside the window (not necessarily consecutive)', - f_host: 'Bind address', - h_host: '127.0.0.1 keeps it local/reverse-proxy only; 0.0.0.0 exposes the port (auth still always on)', - f_port: 'Port', - h_port: 'Production uses 8808 and Nginx points at it — change the proxy too', - f_timeout: 'Upstream timeout (s)', - h_timeout: 'Per upstream request timeout', - f_retry: 'Retries for direct model id', - h_retry: 'Retries for the same model (incl. the first try)', - f_health_auth: 'Require auth on /health', // vue-i18n parses {...} as a placeholder: literal braces must be written {'{'} / {'}'}, // otherwise the whole message fails to compile (and the UI using it renders nothing). // Double quotes: this message needs the {'{'} literal-brace form (braces are vue-i18n placeholders) - h_health_auth: "Off = /health only returns {'{'}\"status\":\"ok\"{'}'} for probes, no dashboard data", - f_hot_reload: 'Config hot reload', - h_hot_reload: 'Reloads hand-edited files by mtime; saving from this page is always immediate', - f_hot_reload_interval: 'Hot reload interval (s)', - h_hot_reload_interval: 'Lower = picks up hand edits sooner, higher = less IO', - f_metrics_enabled: 'Record metrics', - h_metrics_enabled: 'Off = /metrics and dashboard counters stop updating (calls still work)', - f_rate_limit_rps: 'Global QPS per account', - h_rate_limit_rps: 'Empty = unlimited (provider-level limits live in the provider entry)', - f_max_body_size: 'Max request body (bytes)', - h_max_body_size: '~2MB = 2000000; keep it generous if you send images', - f_cors_origins: 'CORS allowed origins', - h_cors_origins: '"*" lets any website drive this gateway from a browser (spending your quota); use trusted domains in production', - f_graceful_shutdown: 'Graceful shutdown wait (s)', - h_graceful_shutdown: 'Seconds to wait for in-flight requests after a stop signal', - f_batch_max_concurrency: '/batch max concurrency', - h_batch_max_concurrency: 'Larger client values are clamped to this', - f_batch_max_items: '/batch max items', - h_batch_max_items: 'Beyond this → 400', - f_webhook_allow_private: 'Allow private webhook callbacks', - h_webhook_allow_private: 'SSRF guard, off by default; enables 192.168./10./loopback callback URLs', - f_docs_enabled: 'Expose /docs', - h_docs_enabled: 'FastAPI built-in API docs — off by default (it lists your endpoints)', - f_auth_fail_limit_per_min: 'Auth failures per minute', - h_auth_fail_limit_per_min: 'Per source IP; beyond this → 429 (brute-force guard)', - f_session_ttl_minutes: 'Web session TTL (minutes)', - h_session_ttl_minutes: 'Login session lifetime, default 1440 = 24h; affects new logins only', - pKey: 'Internal key', - h_pKey: 'Identifier in config (also the /health provider key); renaming creates a new one', - pName: 'Display name', - pBaseUrl: 'Base URL', - pTokenEnv: 'Key env var', - h_pTokenEnv: 'The secret lives in .env; put the variable name here (e.g. AGNES_API_KEY) instead of plaintext', - pToken: 'Plaintext key', - h_pToken: 'Only if you do not use an env var; after saving it is never echoed back', - pRateLimit: 'Provider QPS limit', - h_pRateLimit: 'Empty = unlimited (shared by all models of this provider)', - pHeaders: 'Extra headers', - h_pHeaders: 'Some providers need specific headers (auth/version); leave empty if not', - pMedia: 'Image / video endpoints', - h_pMedia: 'Only for non-standard endpoints (Agnes uses its own paths); all empty = no image/video', - pModels: 'Models', - h_pModels: 'Model ids must match upstream; vision = accepts image input (preferred for image chats)', - mVision: 'vision', - f_o_log_level: 'Log level', - h_o_log_level: 'DEBUG is noisy; INFO is fine for production', - f_o_log_json: 'Structured JSON logs', - h_o_log_json: 'Enable for ELK / CloudWatch; plain text is nicer for humans', - f_o_access_log_maxlen: 'Access log capacity', - h_o_access_log_maxlen: 'Upper bound of "loaded N entries" on the logs page (in-memory ring buffer)', - pricing: 'Model pricing', - h_pricing: 'CNY per million tokens, used for estimated_cost_cny; unset = omitted (calls unaffected)', - priceIn: 'Input price', - priceOut: 'Output price', - dlHosts: 'Allowed download hosts', - h_dlHosts: 'Artifact CDN domains (incl. subdomains); empty = download proxy disabled. Add an entry when the CDN changes', - f_max_bytes: 'Max file size (bytes)', - h_max_bytes: 'Default 67108864 = 64MB', }, // Round 30: user-visible errors from the frontend upload/compression helpers diff --git a/frontend/src/locales/zh.ts b/frontend/src/locales/zh.ts index dd93669..4498895 100644 --- a/frontend/src/locales/zh.ts +++ b/frontend/src/locales/zh.ts @@ -21,10 +21,10 @@ export default { more: '更多', subdashboard: '网关运行概况 · 数据每 8 秒刷新', subchat: '支持附图视觉 · 多供应商直连', - subimages: '参数在左 · 历史在右 · 下载走网关代拉', + subimages: '参数在左 · 历史在右 · 下载更省心', subvideos: '720P · 时长 4–12 秒 · 异步轮询', subagents: '能干活的助手,可按需创建', - submodels: '供应商 / 模型 / 协议线', + submodels: '供应商 / 模型 / 格式', submetrics: '用量与成本 · 按账号隔离', subkeys: '每个账号最多 3 把 · 明文保存,可反复复制', subusers: '账号与权限 · 管理员可查他人日志与资产', @@ -93,8 +93,8 @@ export default { provShare: '供应商成功调用份额', latency: '模型延迟 P50 / P95(s)', providerStatus: '供应商状态', - mediaCircuit: '媒体熔断', - circuit: '熔断', + mediaCircuit: '媒体服务', + circuit: '自动保护', modelCountShort: '模型数', keySet: '已配 Key', name: '名称', @@ -103,12 +103,12 @@ export default { running: '运行中', uptime: '已运行', gwVersion: '网关版本', - concurrency: '当前并发 in-flight {cur} / 峰值 {peak}', + concurrency: '当前并发 {cur} / 峰值 {peak}', latTitle: '模型延迟 P50 / P95(s)', circ_closed: '正常', - circ_half_open: '半开', - circ_open: '熔断', - circ_not_tracked: '未跟踪', + circ_half_open: '恢复中', + circ_open: '已暂停', + circ_not_tracked: '未监控', noData: '暂无数据', requests: '请求总数', successRate: '成功率', @@ -125,7 +125,7 @@ export default { noModels: '暂无调用记录', share: '份额', live: '实时', - byProtocol: '协议线覆盖', + byProtocol: '格式覆盖', byProtocolSub: '每条线有多少模型', modelCountN: '{n} 个模型', }, @@ -142,17 +142,17 @@ export default { fail: '失败', ok: 'ok', none: '暂无', - circuitStatus: '熔断器状态', + circuitStatus: '供应商自动保护', provider: '供应商', state: '状态', recentFailures: '近期失败', - window: '窗口(件)', - threshold: '阈值', - cooldown: '熔断冷却(s)', - probing: '探测中', - stateClosed: 'CLOSED', - stateHalf: 'HALF_OPEN', - stateOpen: 'OPEN', + window: '统计次数', + threshold: '触发次数', + cooldown: '暂停时长(s)', + probing: '检测中', + stateClosed: '正常', + stateHalf: '恢复中', + stateOpen: '已暂停', loadErr: '加载指标失败', requests: '请求总数', failCount: '失败数', @@ -165,7 +165,7 @@ export default { inflightPeak: '在途 / 峰值', failRateNote: '失败率 {v}%', snapshot: '实时快照', - circuitCard: '熔断器', + circuitCard: '连续失败会自动暂停,稍后自动恢复', byProvider: '按供应商调用(成功 / 失败)', }, logs: { @@ -238,7 +238,7 @@ export default { params: '参数', clear: '清空', generating: '正在生成…', - empty: '开始与网关对话(在下方选一个模型)', + empty: '开始对话(在下方选一个模型)', servedBy: '由 {provider} 响应', requestedTo: '请求 {req} → 实际 {served}', reasoning: '推理过程(reasoning)', @@ -253,7 +253,7 @@ export default { imgMax: '一条消息最多 {n} 张图片', imgPickErr: '图片处理失败:{msg}', visionWarn: '当前模型 {model} 不支持图像输入,图片会被忽略或直接报错 —— 请换用带视觉能力的模型。', - paramLocked: '该模型参数由网关固定:{p}(上游硬约束,你在这里调的值会被网关覆盖后再发送)', + paramLocked: '该模型参数已固定:{p}(你在这里调的值会被覆盖后再发送)', useVision: '改用 {model}', waiting: '等待输入…', send: '发送', @@ -282,11 +282,11 @@ export default { refPickErr: '图片处理失败:{msg}', refsPH: '图生图 / 多图合成:图片 URL 或 data:,每行一个(留空 = 纯文生图)', count: '张数', - countTip: '上游一次只出 1 张,这里会逐张调用', + countTip: '一次只出 1 张,这里会逐张调用', genProgress: '正在逐张生成 {i}/{n}', partialFail: '已生成 {done} 张,第 {i} 张失败:{msg}', refsUsed: '{n} 张参考图', - emptyResp: '上游没有返回图片', + emptyResp: '没有返回图片', placeholder: '描述想生成的图像(可给风格/主体/构图)', servedBy: '由 {provider} 生成', clear: '清空历史', @@ -317,7 +317,7 @@ export default { uploadAudio: '上传参考音频', useUrl: '用 URL', urlPH: '或粘贴图片 URL', - uploadTip: '上传的素材会在浏览器内压缩为 data URI 直传上游(无需公网可访问的 URL)', + uploadTip: '上传的素材会在浏览器内压缩后直接发送(无需公网可访问的 URL)', pickErr: '素材处理失败:{msg}', refMax: '参考图最多 {n} 张', audMax: '参考音频最多 {n} 段', @@ -371,14 +371,13 @@ export default { agents: { title: '智能体', sub: '能干活的助手 —— 每个智能体有自己的参数和会话', - subHint: '{n} 个智能体 · 引擎 {engine}({url})· {scope}', scopeAdmin: '仅管理员可见', scopeOpen: '{n} 人可用', new: '新建', newAgent: '新建智能体', empty: '还没有智能体', emptyPick: '从左侧选一个智能体', - emptyHint: '或新建一个 —— 它就是「带模型的助手」:引擎 + 参数 + 独立会话', + emptyHint: '或新建一个 —— 它就是「带模型的助手」:模型 + 参数 + 独立会话', tabChat: '对话', tabConfig: '配置', tabRuns: '运行记录', @@ -391,26 +390,26 @@ export default { // 对话 newChat: '新建对话', thisChat: '本条对话', - engineReady: '引擎就绪', - engineDown: '引擎不可达', - engineDownHint: '引擎未就绪,请稍后再试', + engineReady: '可用', + engineDown: '暂不可用', + engineDownHint: '暂时连不上,请稍后再试', retry: '重试连接', stop: '停止', stopRun: '转后台', running: '生成中', - inputPlaceholder: '说点什么… 这个智能体能读写文件、跑命令、开浏览器(都在服务器上)', + inputPlaceholder: '说点什么… 这个智能体能读写文件、跑命令、开浏览器', execSteps: '执行了 {n} 步', execRunning: '执行中', tokens: 'Tokens', elapsed: '耗时', toolsUsed: '工具', - interrupted: '⚠ 引擎连接中断,本轮未完成', + interrupted: '⚠ 连接中断,本轮未完成', // 配置 cfgBasic: '基本', - cfgBasicSub: '显示名与 slug 可改', + cfgBasicSub: '显示名与标识可改', cfgName: '名称', - cfgSlug: 'slug', - cfgSlugHint: '不能与已有的模型 id 或别名重名', + cfgSlug: '标识', + cfgSlugHint: '不能与已有的模型名重复', cfgDesc: '描述', cfgModel: '模型', cfgModelPick: '选择模型', @@ -418,8 +417,9 @@ export default { cfgModelSwitched: '已切换到 {model}', cfgModelAtChat: '在对话框左下角切换(所有智能体共用,即时生效)', cfgExec: '执行', - cfgExecSub: '影响服务器安全,一期只读', - cfgEngine: '引擎', + cfgExecSub: '涉及运行安全,一期只读', + cfgEngine: '运行环境', + cfgEngineVal: '由平台提供', cfgWorkspace: '工作区', cfgParams: '参数', cfgTemp: 'temperature', @@ -443,7 +443,7 @@ export default { limitReached: '已达上限({n} 个)', limitHint: '已达上限 {n} 个(软上限);需要更多时先归档不用的', sharedNote: '每个智能体有自己的参数和会话,模型是共用的', - riskNote: '该智能体可在服务器上读写文件、执行命令', + riskNote: '该智能体可读写文件、执行命令', // 运行记录 runsTotal: '调用次数', runsTools: '工具调用', @@ -474,7 +474,7 @@ export default { runsEmpty: '还没有运行记录', // 表单 formName: '名称', - formSlug: 'slug', + formSlug: '标识', formDesc: '描述', formTemp: 'temperature(可选)', formSuffix: '角色补充(可选)', @@ -507,7 +507,7 @@ export default { freeLimitedTag: '限时免费', freeLimitedHint: '官方「限时免费」:优惠有期限,官方改口径后每日同步会自动更新这个标注', paramTag: '参数固定', - paramLocked: '该模型参数由网关固定:{p}(上游硬约束,客户端与测速探针传的值都会被覆盖)', + paramLocked: '该模型参数已固定:{p}(客户端与测速传的值都会被覆盖)', allTextModels: '全部文本模型', providerCard: '供应商', keyOk: 'Key 已配置', @@ -522,19 +522,19 @@ export default { failed: '失败', speedtestDone: '测速完成,共 {n} 个模型', speedtestFail: '测速失败', - speedtestHint: '按模型自己的协议线探测延迟', + speedtestHint: '按模型自己的格式探测延迟', textModelCount: '{n} 个文本模型', // 第二十七轮:支持的格式列(三条协议线)。格式名直接用各家标准端点名,不译。 protocols: '支持的格式', - protocolHint: '该模型可以走哪几条线:chat/completions(OpenAI 兼容)、messages(Anthropic)、responses(OpenAI Responses)。不在列表里的格式会在网关入口直接 400 —— 网关不做跨格式转换,也绝不偷偷换个模型顶上', + protocolHint: '该模型支持哪几种调用格式:chat/completions(OpenAI 兼容)、messages(Anthropic)、responses(OpenAI Responses)。用不在列表里的格式会直接报 400 —— 不会自动转换格式,也不会偷偷换个模型顶上', protocolNa: '不适用', - protocolNaHint: '图像 / 视频模型走 /v1/images/generations、/v1/videos 专用端点,不涉及这三条协议线', + protocolNaHint: '图像 / 视频模型走 /v1/images/generations、/v1/videos 专用地址,不用上面这三种格式', protocolHintRow: '支持的格式:{lines} ', probedLine: '本次测速走 {p}', // 第三十一轮:虚拟模型 id(别名)—— 客户端固定写 model: wjd,实际转发到管理员选的模型。 // 别名不是模型:不进测速列表、不进模型计数,只有这一张卡片。 aliasCard: '虚拟模型 id(别名)', - aliasLine: '只支持 chat/completions(OpenAI)格式:用 /v1/messages、/v1/responses 打 {id} 会在入口直接 400(网关不做跨格式转换)。', + aliasLine: '只支持 chat/completions(OpenAI)格式:用 /v1/messages、/v1/responses 调 {id} 会直接报 400(不会自动转换格式)。', aliasInherit: '视觉能力、强制参数与定价都跟随目标模型 —— 别名不另存一份配置,改目标即可整体切换。', aliasName: '别名 ID', aliasTarget: '目标模型', @@ -667,7 +667,7 @@ export default { epDownload: '代下载生成出来的图片 / 视频(绕过上游地址的防盗链与过期,白名单域名)', epHistory: '读取本人历史(图像 / 视频 / 对话)', epLogs: '本人调用日志', - epMetrics: '指标快照(含熔断器状态)', + epMetrics: '指标快照(含供应商保护状态)', epHealth: '健康检查(版本 / 供应商状态)', exChat: '示例 · 对话', exMessages: '示例 · 对话(Anthropic Messages 格式)', @@ -767,25 +767,15 @@ export default { file: '网关配置文件', fileSub: '管理员专属 · 保存即生效', path: '文件位置', - restartKeys: '这几项改了要重启进程', + restartKeys: '这几项改了要重启才生效', otherKeys: '其余所有项', - otherKeysTip: '每请求读取,保存后立刻生效(无需重启、无需等热重载)', - saveTip: '保存会用页面上的值重写整个配置文件(旧文件先备份为 .bak,可回滚);密钥只显示"配没配",不会回显到浏览器。', - viewYaml: '查看配置文件', + otherKeysTip: '保存后立刻生效,不用重启', + saveTip: '保存会用页面上的值重写整个配置文件(旧文件会自动备份一份,可回滚);密钥只显示"配没配",不会回显到浏览器。', save: '保存并生效', - saved: '已保存并生效', - discard: '放弃改动并重载', editorTitle: '配置文件内容(gateway.yaml)', - editorSub: '直接编辑完整 YAML,点「保存」即应用', - exportCfg: '导出配置', - importCfg: '导入配置', - importCfgOk: '配置已导入编辑器,确认无误后点「保存」应用', + editorSub: '直接编辑完整配置内容,点「保存」即应用', envTitle: 'API Key 环境变量', - envSub: '存于 data/.env;默认隐藏,可显示、复制、修改', - envExport: '导出变量', - envImport: '导入变量', - envImportErr: '变量导入失败', - envEmpty: '暂无使用 token_env 的供应商', + envSub: '存在 data/.env 文件里;默认隐藏,可显示、复制、修改', envSet: '已配置', envMissing: '未配置', envShow: '显示', @@ -794,7 +784,7 @@ export default { envCopied: '已复制', envSaved: '已保存到环境变量并生效', saveOk: '已保存并生效', - restartPill: '需重启服务', + restartPill: '需重启才生效', restartSaved: '本次改动了需重启才能生效的项:{keys},重启后生效', backup: '备份配置', restore: '恢复配置', @@ -806,124 +796,6 @@ export default { restoreInvalid: '备份文件无效', errTitle: '保存失败,配置未被改动', - needRestart: '需重启', - restartApplied: '已保存。这几项要重启 api-gateway 才生效:{keys}', - savedRestart: '已保存并生效', - addItem: '添加', - addModel: '添加模型', - addProvider: '添加供应商', - addPrice: '添加定价', - delete: '删除', - delProvider: '删除供应商', - delConfirm: '删除「{name}」?保存后才真正写盘生效。', - needKey: '供应商的 key 不能为空', - dupKey: '「{name}」重复了,请改一个名字', - newProvider: '(新供应商)', - unlimited: '留空 = 不限', - useServerTimeout: '留空 = 用 server.timeout', - mediaEmpty: '留空 = 不支持', - modelId: '模型 id', - modelDisplay: '显示名', - hdrKey: '请求头', - hdrVal: '值', - tokenSet: '已配置密钥', - tokenEnvMissing: '环境变量未设置', - tokenKeep: '留空 = 不修改(当前已配置)', - tokenEmpty: '留空 = 不设置', - jump: '跳到', - tabServer: '服务与安全', - tabServerSub: '监听、超时/重试、限流、CORS、会话时长等运行开关', - tabProviders: '供应商与模型', - tabProvidersSub: '供应商密钥来源、接口地址、图像/视频端点、模型清单', - tabLog: '日志与定价', - tabLogSub: '日志级别/格式/容量 + 模型定价(人民币 / 百万 token)', - tabDownload: '下载代理', - tabDownloadSub: '允许网关代下载的 CDN 域名与单文件上限', - circuit: '供应商熔断', - h_circuit: '某供应商窗口内失败到阈值就熔断冷却,期间它的请求直接 503', - f_c_enabled: '启用熔断', - f_c_failure_threshold: '触发阈值(次)', - f_c_cooldown: '熔断冷却(秒)', - f_c_window: '统计窗口(秒)', - h_c_enabled: '关掉 = 失败也不跳过', - h_c_failure_threshold: '窗口内失败到该次数就熔断', - h_c_cooldown: '熔断期间直接跳过该供应商,冷却结束后自动半开试探', - h_c_window: '按"窗口内失败次数"计数(不是连续失败)', - // ── server 段(键名与 gateway.yaml 一致,页面标签 = f_)── - f_host: '监听地址', - h_host: '127.0.0.1 只对本机/反代开放;改成 0.0.0.0 会对外暴露端口(鉴权仍恒定开启)', - f_port: '监听端口', - h_port: '生产用 8808,且 Nginx 反代指向它 —— 改了要同步改反代', - f_timeout: '上游超时(秒)', - h_timeout: '单次上游请求超时', - f_retry: '直连模型重试次数', - h_retry: '同一模型的重试次数(含首次)', - f_health_auth: '/health 需要鉴权', - // 注意:vue-i18n 把 {...} 当占位符,花括号要么写成 {'{'} / {'}'},要么整条编译不过 —— - // 编译不过的表现是"用到这条文案的整块界面不渲染",不报在文案上(踩过一次,见 check:i18n)。 - // 用双引号是因为这条文案里有 {'{'} 这种字面量花括号写法(花括号是 vue-i18n 的占位符语法) - h_health_auth: "关着时 /health 只回 {'{'}\"status\":\"ok\"{'}'}(供探活),不回看板数据", - f_hot_reload: '配置热重载', - h_hot_reload: '按 mtime 自动重载手改的文件;页面上保存本来就是即时生效,与此无关', - f_hot_reload_interval: '热重载检查间隔(秒)', - h_hot_reload_interval: '越小改文件越快被发现,越大越省 IO', - f_metrics_enabled: '记录调用指标', - h_metrics_enabled: '关掉后 /metrics 与看板计数停更(不影响调用)', - f_rate_limit_rps: '全局单账号 QPS 上限', - h_rate_limit_rps: '留空 = 不限(供应商级限流见供应商配置)', - f_max_body_size: '请求体上限(字节)', - h_max_body_size: '约 2MB = 2000000;带图对话别调太小', - f_cors_origins: 'CORS 允许来源', - h_cors_origins: '填 * 表示任意站点都能跨域调本网关(会消耗你的额度),生产建议换成可信域名', - f_graceful_shutdown: '优雅关闭等待(秒)', - h_graceful_shutdown: '收到停止信号后等在途请求的秒数', - f_batch_max_concurrency: '/batch 并发上限', - h_batch_max_concurrency: '客户端传更大的值会被钳到这里', - f_batch_max_items: '/batch 单次条目上限', - h_batch_max_items: '超出直接 400', - f_webhook_allow_private: '允许视频回调指向内网', - h_webhook_allow_private: 'SSRF 防护开关,默认关;开了才能回调 192.168./10./回环地址', - f_docs_enabled: '开放 /docs 文档', - h_docs_enabled: 'FastAPI 自带的接口文档,安全默认关(会暴露接口清单)', - f_auth_fail_limit_per_min: '鉴权失败限流(次/分)', - h_auth_fail_limit_per_min: '同一来源 IP 每分钟鉴权失败超过该值 → 429,防暴力枚举', - f_session_ttl_minutes: '网页会话时长(分钟)', - h_session_ttl_minutes: '登录签发的会话有效期,默认 1440 = 24 小时;只影响新登录', - // ── 供应商 ── - pKey: '内部键名', - h_pKey: '配置里的标识(也是 /health 里的 provider key),改名等于新增一个', - pName: '显示名', - pBaseUrl: '接口地址 base_url', - pTokenEnv: '密钥环境变量', - h_pTokenEnv: '密钥存在 .env 里,这里填变量名(如 AGNES_API_KEY),不把明文写进配置', - pToken: '明文密钥', - h_pToken: '没有用环境变量时才填;保存后只回报"已配置",不会再回显', - pRateLimit: '本供应商 QPS 上限', - h_pRateLimit: '留空 = 不限(供应商维度,所有模型共享)', - pHeaders: '额外请求头', - h_pHeaders: '少数供应商要带特定头(如自定义鉴权/版本头),没有就留空', - pMedia: '图像 / 视频端点', - h_pMedia: '非标准端点的供应商才填(Agnes 是自有路径);全留空 = 该供应商不支持图像/视频', - pModels: '模型清单', - h_pModels: '模型 id 必须与上游一致;vision = 支持图像输入(带图对话会优先选它)', - mVision: '视觉', - // ── 可观测性 ── - f_o_log_level: '日志级别', - h_o_log_level: 'DEBUG 最啰嗦,生产 INFO 够用', - f_o_log_json: '结构化 JSON 日志', - h_o_log_json: '接 ELK / CloudWatch 时开;人看还是普通格式舒服', - f_o_access_log_maxlen: '访问日志条数上限', - h_o_access_log_maxlen: '日志页「已加载 N 条」的上限(内存环形缓冲)', - pricing: '模型定价', - h_pricing: '人民币 / 百万 token,用于返回 estimated_cost_cny;不填则该项不返回(不影响调用)', - priceIn: '输入价', - priceOut: '输出价', - // ── 定价 ── - // ── 下载代理 ── - dlHosts: '允许代下载的域名', - h_dlHosts: '产物 CDN 域名(含子域);留空 = 整个下载代理关闭。换了 CDN 就往这里加一条', - f_max_bytes: '单文件上限(字节)', - h_max_bytes: '默认 67108864 = 64MB', }, // 第三十轮: 前端上传/压缩工具的用户可见报错(此前硬编码中文 —— 切英文后 toast 里还是中文) diff --git a/frontend/src/views/Agents.vue b/frontend/src/views/Agents.vue index 917f000..1bb40eb 100644 --- a/frontend/src/views/Agents.vue +++ b/frontend/src/views/Agents.vue @@ -312,9 +312,7 @@ {{ t('agents.cfgExecSub') }}
{{ t('agents.cfgEngine') }}
-
{{ cur.engine?.type }} · {{ cur.engine?.base_url }}
-
{{ t('agents.cfgWorkspace') }}
-
{{ cur.engine?.agent_id }}
+
{{ t('agents.cfgEngineVal') }}