Commit Graph
15 Commits
Author SHA1 Message Date
rootandClaude Opus 5.5 1fcfcac885 feat(browser): keep sessions alive during desktop, takeover and agent use (AG-4)
camofox reaps a tab after TAB_INACTIVITY_MS (5 min) without API tool calls;
desktop-mode input goes straight to Selkies and screenshots don't count, so
tabs disappeared under the user during takeover.

- BrowserKeepalive (in-process): sessions held by an open desktop WebSocket
  (ref-counted), a human takeover lease, or a running/queued agent run that
  used browser.* get a keepalive every BROWSER_KEEPALIVE_MS (default 120s,
  0 = off). Holds are re-checked each round; release/lease expiry/agent end/
  session close stop it and camofox reaps as usual.
- Keepalive call: camofox 1.17 has no keepalive/touch endpoint and /stats does
  not bump toolCalls; use GET /tabs/{id}/downloads without includeData/consume
  (read-only metadata). Goes through BrowserService.run so restarts or reaped
  tabs self-heal via BR-3 reconnect; no lastAutomationAt, no control needed.
- Runbook browser.md §12, deploy env table, compose env; tests against the
  testkit mock camofox.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-30 18:05:12 +08:00
rootandClaude Opus 5.5 adc853aa64 feat(sites): site requests follow browser UA, smooth UA switch, CF cookies browser-only (BR-4)
- every outbound site request (adapter list/search/detail/download/health,
  RSS, login, user data) carries the account UA via SourcesRuntime.siteFetch;
  no more Deno default UA
- priority: account manual > pending > global setting > browser identity
  (fallback: same-shape Windows Firefox UA); settings UA default is empty
- migration 0056: existing cookie/password accounts keep their old UA as
  ua_pending until re-login or browser cookie write-back switches cookie+UA
  in one transaction; browser_provider.identity caches the browser identity
- cf_clearance/__cf_bm stripped on browser write-back (names kept only)
- browser sessions get per-site egress (same proxy as HTTP, or direct)
- site account page shows UA source; raw UA in tech info

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-30 03:17:13 +08:00
rootandClaude Opus 5.5 fa344a7184 feat(web): token 数以 K 输入显示,已存凭证显示掩码占位(FIX-8)
- 新增 UnitInput 与 kind=tokens:上下文预算、单次 token 上限、自定义模型上下文窗口以 K 输入(允许小数),存储仍为原始 token 数;校验与提示用 K 表述;列表/详情 token 数统一 K/M 简写(formatTokens)
- 新增 SecretInput 与 kind=secret:已保存时显示「已设置 ••••••••」(有服务端掩码时带末 4 位),聚焦直接输入替换,留空保持,「清除」确认后保存生效;替换模型服务、TMDB、网页搜索、下载器、媒体服务器、浏览器服务、站点账号、Passkey、本机 Token 各处
- 统一只写凭证服务端语义:省略或空串 = 不变,null = 清除(contracts secretInput);下载器/媒体服务器新增清除;站点凭证同类更新可留空保持密码/Cookie/API Key/TOTP,TOTP 可 null 清除,不再误删 TOTP 与 UA
- contextBudget 最小值 4096→4000、contextWindow 1024→1000,便于按 K 填写
- 测试:换算与校验、SecretInput 渲染与状态、各接口留空不变/显式清除

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-30 00:10:46 +08:00
rootandClaude Opus 5.5 4480fc346d refactor(copy): 术语统一——收录版本、模型服务、智能体等(DEAI-2)
- 版本目标 → 收录版本(前后端用户可见字符串、能力描述、给模型的提示)
- Agent 上游 / 上游 → 模型服务;界面上的 Agent / Agent 会话 → 智能体 / 智能体会话
- 身份歧义 / 识别有歧义 / 无法确定作品 → 没识别出作品;待识别(池) → 未识别
- 建档 → 加入媒体库;资产 → 文件;库根 → 媒体库目录;纳管 → 本工具 / 外部任务;
  哨兵 → 标记文件 / 检查目录;站点快照 → 抓取;Feed / RSS 订阅 / 来源健康 → RSS
- 代码标识符、API 字段名、迁移文件不改;glossary.md 同步;smoke 选择器跟随

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 20:35:43 +08:00
root f41cd96fee Merge branch 'hh2'
# Conflicts:
#	apps/server/src/modules/browser/service.ts
#	apps/web/app/components/browser/desktop-view.tsx
#	apps/web/app/components/browser/site-origin-view.tsx
#	packages/contracts/src/browser.ts
2026-09-29 17:43:00 +08:00
rootandClaude Opus 5.5 cabd772243 fix(sites): normalize HHClub RSS items, complete detail page parsing (HH-2)
- RSS: split NexusPHP "[category]name[subtitle][size][uploader]" titles in the
  site adapter layer (category id/name/media type, clean title, subtitle,
  size); snapshot fields win on merge; lazily renormalize old RSS rows and
  re-emit release.observed; unknown list fields show an em dash
- Detail: HHClub section layout (douban info block as description,
  screenshots section, IMDb/TMDB from MediaInfo lines, upload time field,
  on-demand viewfilelist.php once); site CDN screenshots get a Referer
- Site request count now includes site-bound RSS fetches
- Origin view shows "自动化" when idle and "Agent" only after recent
  automation actions (lastAutomationAt); agent step titles no longer get
  squeezed to a sliver by long notes

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 17:30:25 +08:00
rootandClaude Opus 5.5 838aea3ebd feat(browser): self-heal sessions after camofox restarts (BR-3)
- Any session operation that hits 404/410 (or finds no open tab) reconciles
  tabs; if none survive it re-injects the account cookies (BR-2), opens a new
  tab under the same userId/sessionKey, returns to the recorded URL (site home
  otherwise), marks old tabs lost and retries the operation once. Failure
  returns 503 browser_reconnect_failed with the reason, with a 10s backoff.
- Restart detection: umt plugin GET /umt/instance (fallback /health fields);
  a changed instance id marks every open session needsReconnect.
- POST /browser/sessions/{id}/reconnect (idempotent); session view exposes
  needsReconnect and reconnect {status, trigger, at, reason, humanNotice}.
- Controller and lease are kept; reconnecting during takeover asks the human
  to double-check the reloaded page.
- Web: screenshot and desktop modes show "reconnecting", auto-reconnect once,
  pause polling, offer a reconnect button with the reason on failure; desktop
  mode rebuilds the Selkies connection after a reconnect.
- Mock camofox restart(), tests, smoke step and runbook section 10.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 17:29:11 +08:00
rootandClaude Opus 5.5 28974d1e1a feat(browser): unify site credentials with browser profiles (BR-2)
- Inject the site account Cookie (manual + auto-login session, session wins)
  into the camofox profile before the first tab opens; password accounts
  without a session run refresh-login first; API Key sites are skipped.
- Cookie expired after injection -> credential.status=expired; cookie
  accounts go to waiting/login_required ("log in once, applies site-wide").
- Write back site-domain cookies from the browser on handover, session close,
  every 10 min of activity, and from refresh-login when the stored cookie is
  invalid (verified before saving). user_info.cookieOrigin records
  source=browser; source shown as manual/login/browser in the origin view.
- New camofox plugin umt-cookies (GET/POST /umt/sessions/:userId/cookies
  under the access key); profile dir 0700/files 0600 in the desktop image.
- Single entry: sources-identity/service/credentials.ts.
- Real check: HHClub cookie injected into a fresh profile -> home page
  logged in (logout link present), 5 cookies read back.

Requires rebuilding umt-browser-desktop (docker compose build camofox &&
docker compose up -d camofox) to take effect.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 15:25:07 +08:00
rootandClaude Opus 5.5 c115676e23 fix(browser): opening a session ignores ones another process already closed
The worker's adapter still held a session the api process had closed,
so browser.navigate {siteId} from an agent run failed with "该站点/账号
已有未关闭的浏览器会话" (REAL-2 task 4). When the database has no live
session for the profile, stale in-memory ones are marked closed before
opening. Needs image rebuild.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 03:15:46 +08:00
rootandClaude Opus 5.5 e53d05a872 fix(browser): reload session state when another process switched the tab
api and worker processes each keep adapter state in memory. After the
upstream reaped an idle tab and the api process reopened one, the agent
worker kept pointing at the lost tab and every screenshot / snapshot in
solve_challenge failed with not_found (REAL-2 task 4). loaded() now
restores from the database when its current tab differs. Needs image
rebuild.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 03:13:40 +08:00
rootandClaude Opus 5.5 c437c6bff6 feat(browser): vision automation loop with screenshots, coordinate actions and bounded challenge solving
- browser.screenshot returns the page as an image content block (≤1280px wide, JPEG quality
  configurable) for vision-capable upstreams; non-vision upstreams get a redacted text snapshot.
  Login pages / password fields are marked sensitive and withheld unless a
  browser.sensitive_screenshot grant exists and allowSensitive is passed.
- browser.click_at / type_at / scroll / wait act in screenshot pixels; the server converts via
  viewport/DPR, requires automation control and an allowlisted host, and writes a thumbnail
  evidence file (DATA_DIR/tmp/evidence) referenced from agent_run_event.
- browser.solve_challenge runs a bounded child agent_run (vision caps only, maxSteps ≤ 12,
  budget), judges success from page facts, stops immediately on 2FA / SMS / device confirmation,
  and otherwise moves the session to waiting with evidence and a takeover link.
- PiAgentRuntime passes modelImage outputs as pi-ai image content when model.input has image;
  the ledger never stores image data; coordinate actions use per-invocation idempotency keys.
- Tests: image codec, runtime image content, E14 e2e (login solved, 2FA stop, step limit,
  sensitive withheld). Real camofox 1.17.0 check on example.com recorded in the runbook.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:27:48 +08:00
rootandClaude Opus 5.5 40dd0e69e4 feat(browser): ship camofox v1.17.0 in compose, verified on a real instance
- compose: camofox service on by default (internal only, volume umt-camofox,
  healthcheck, CAMOFOX_REPLICAS=0 to skip); app depends on it with
  service_started/required=false and auto-registers camofox-env from
  BROWSER_PROVIDER_URL / BROWSER_ACCESS_KEY / BROWSER_ENABLED
- docker/camofox/Dockerfile: builds upstream tag v1.17.0 (389c996); fixes the
  upstream Dockerfile (missing postinstall.js before npm ci, sh -c not
  forwarding SIGTERM so logins were never persisted on restart)
- adapter fixes from the real container: coordinate click via evaluate +
  selector, focus typing with mode=keyboard, password values redacted from
  ARIA snapshots/extract, downloads field names, 410 after browser restart
  treated as lost tab; navigate reopens a tab in the same session
- verified-capability table keyed by upstream version (1.17.0 all ten);
  other versions stay unverified; probe reports auth failures
- settings: browser providers block (create/edit/delete/probe, capability
  table); API POST/PATCH/DELETE /browser/providers, POST .../probe
- scripts/browser-verify: umtbr stack + mock NexusPHP run of REQ 4.7.4 items
- docs: runbook browser §1/§7 (real-instance record), deploy §2.1

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 00:05:21 +08:00
rootandClaude Opus 5.5 b4c9b1ead2 fix(system): retired heartbeat rows no longer alarm on status page
- GET /system/health splits processes into current instances (hosts with
  a beat in the last 10 min, newest per host+kind, not stopped) and
  retiredProcesses[]; 45s stale and level only consider current ones
- heartbeat writer marks stopped_at on graceful exit (migration 0043),
  upsert clears it; pure api role now stops its writer on SIGTERM
- scheduled job system.heartbeat_cleanup (maintain, 1h) deletes rows
  without a beat for 24h and samples older than 7 days
- status page: per-role alarms ("io worker 45 秒没有心跳"), collapsible
  retired instances; glossary/runbook updated
- bump all members to 0.2.2, CHANGELOG [0.2.2]

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 23:31:21 +08:00
rootandClaude Opus 5.5 fb84a71f80 style(copy): fix remaining punctuation violations and make check:copy strict
Auto-fix of 602 string literals across 74 files (half-width punctuation next to
Chinese, parenthesised Chinese, quotes); biome reflow; e2/e6 assertions follow
the new full-width summaries. check:copy now runs with --strict in ci.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:34:51 +08:00
rootandClaude Opus 5.5 adffe36846 feat(server): browser module with DB-backed sessions, takeover lease and browser.* capabilities
/browser/providers, /browser/sessions (open/close/list), navigate/snapshot/screenshot/
click/type/scroll/select/extract/download and control acquire/release. Human input
needs an active lease; expiry or captcha moves the session to waiting with a decision;
release forces a fresh snapshot. Downloads are staged under DATA_DIR/tmp, verified by
infohash and submitted through POST /acquisition-plans. Capability count 21 -> 25.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 20:17:46 +08:00