Commit Graph
255 Commits
Author SHA1 Message Date
rootandClaude Opus 5.5 7e028fa23e feat(server): proxy browser desktop with lease-based control/view credentials
/api/v1/browser/desktop/{sessionId}/* proxies Selkies HTTP and WebSocket,
console sessions only. The holder of the active human lease gets the control
password, everyone else the view-only one; the WS bridge rechecks every 3s
and drops the connection on role change so the client reconnects downgraded.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 02:30:55 +08:00
rootandClaude Opus 5.5 880cec4d23 feat(docker): browser desktop image (Selkies Wayland + headed camofox) and optional TLS proxy
Selkies baseimage (debiantrixie, digest-pinned) with camofox 1.17.0 copied
from camofox-browser:1.17.0, run headed under labwc so the desktop shows the
same Firefox the agent drives. s6 service waits for wayland-0 and depends on
svc-de and svc-dbus so shutdown persists login state. Selkies control/view
passwords derive from CAMOFOX_ACCESS_KEY. Compose defaults to the new image;
CAMOFOX_IMAGE=camofox-browser:1.17.0 switches back. Optional caddy (profile tls).

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 02:30:55 +08:00
rootandClaude Opus 5.5 3f0327ad97 test(e2e): AG-3 agent quality with fake model; runbook capability list
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 41e3647f0a feat(web): render agent reports as safe-subset Markdown
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 b7cbc06567 feat(agent): prompt rules for probe queries, param types, Chinese reports and interrupted summaries
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 a00a9eaf5d feat(agent): budget reminders, usage while waiting, budget-exceeded status and dedupe
- tool results carry a budget reminder near limits; text before the last tool
  call is dropped as report when cut by budget/abort
- usage persisted when a run suspends, so waiting sessions show spent tokens
- task page shows budget_exceeded via business.outcome instead of completed
- budget-exceeded decisions merge by root cause (decision / schedule / goal
  template); new summaries added to investigate allowlist

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 4cd9ec8390 fix(pi): upstream completion test sends a sessionId (OpenCode x-opencode-session)
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 975a14ace7 feat(agent): subscription.list, decision.list, download.inspect; grant-scoped requiresApproval
- three read-only summary capabilities (38 registered); task.inspect points
  download task ids to download.inspect
- probe queries reject empty / single-letter / wildcard-only terms; externalId
  documented as string; every description says when to use it
- download.plan requiresApproval uses coveringGrant (shared with authorize)

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +08:00
rootandClaude Opus 5.5 3134e3a55e test(contracts): upstream model list columns are db-only in db-sync
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:58:28 +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 0765264a47 feat(pi): fetch and refresh upstream model lists, searchable model picker
- POST /pi/upstreams/{id}/models/refresh calls the provider's model list
  endpoint (OpenAI-shaped /models, Anthropic /v1/models, Google
  /v1beta/models, OpenRouter /api/v1/models; Radius via pi-ai refresh),
  normalizes {id, displayName, contextWindow, supportsVision,
  supportsTools, pricing} and stores it in pi_upstream.models (0044)
- GET .../models returns {source, fetchedAt, models[]}: remote merged
  with the static catalog (remote first); failures (401/network/
  unsupported) fall back to static with a one-line error
- POST /pi/models/probe for trying credentials before creating
- wizard and edit drawer: fetch button, searchable dropdown with
  vision/tool tags, manual id, source line; default upstream needs a model
- tests: mock /models, 401, network failure; smoke step; runbook

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:26:07 +08:00
rootandClaude Fable 5.1 e5abfa6217 docs(plan): AG-3 agent quality items from real-upstream verification
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 01:24:37 +08:00
rootandClaude Opus 5.5 4b6da1343a fix(agent): auto-recover outcome is failed when a recovery action itself fails
When download.recheck / task.retry was executed but errored (e.g. the
downloader was unreachable), applyRecoveryOutcome only looked at
succeeded actions and recorded outcome=not_attempted with the summary
"没有执行恢复动作", which is false. Record outcome=failed and name the
failed action instead. Needs image rebuild to take effect.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:21:37 +08:00
rootandClaude Opus 5.5 57fe4c11d3 fix(downloads): treat missing site/adapter as permanent source failure
torrentSource threw plain Error for a missing site or a site without an
adapter definition, so submitDownload retried to max attempts, left the
plan and slot reserved, and sweepReservedPlans re-enqueued it forever
(agent runs stayed waiting on goal-settled). Throw SiteError
(not_found / unsupported) so the plan fails and the slot is released.
Needs image rebuild to take effect.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 01:13:19 +08:00
rootandClaude Fable 5.1 ce36a42336 docs(plan): upstream model list refresh; real-upstream testing model
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 01:01:13 +08:00
rootandClaude Fable 5.1 ff098734c6 docs(design): Webtop 4.0 same-session browser desktop plan (placement A, vision loop)
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 01:00:26 +08:00
rootandClaude Fable 5.1 028f0d238e docs(plan): webtop placement comparison and vision-driven browser automation loop
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 00:51:31 +08:00
rootandClaude Fable 5.1 b0a19c36b4 docs(plan): remote browser view via Webtop 4.0 (Wayland/Selkies) with CJK/IME/clipboard
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 00:44:39 +08:00
rootandClaude Opus 5.5 39f5897df6 fix(web): subscribe from a TMDB candidate by creating the entity first
The search page 订阅 button and the new-subscription wizard threw away
the TMDB id and created a title-only subscription, which stays pending
(metadata_pending) with no seasons, episode totals or coverage matrix
until some release happens to be identified. Call
/files/entity-from-candidate first and subscribe the entity; fall back
to title tracking only if that fails. Picking a TMDB row in the wizard
now creates the library entity even if the dialog is then cancelled
(idempotent per TMDB id).

Found in REAL-1b on the v0.2.2 test stack; needs an image rebuild to
take effect.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 00:37:45 +08:00
rootandClaude Opus 5.5 775344f597 fix(identify): rank TMDB candidates by original title too
catalog.lookup only returned the zh-CN localized title, so an English
release name (Dune.Part.Two.2024...) was rejected against "沙丘2" and
files/identify came back with zero candidates. Pass originalTitle
through catalog.lookup and use it as an extra title in ranking.

Found in REAL-1b on the v0.2.2 test stack; needs an image rebuild to
take effect.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 00:33:20 +08:00
rootandClaude Fable 5.1 1de131b192 docs(runbook): note UMT_SUBNET must fall within firewall-allowed range
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 00:30:05 +08:00
rootandClaude Fable 5.1 5f8b933f4b build(compose): pin default network subnet to 172.30.0.0/24 for host firewall rules
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 00:29:06 +08:00
rootandClaude Fable 5.1 1b626f2ee8 docs(runbook): containers via host proxy relay (socat on docker0)
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-29 00:23:59 +08:00
rootandClaude Opus 5.5 d53df25fc8 feat(agent): web.search / web.fetch via Brave Search (WEB-1)
- adapters/websearch: Brave client (X-Subscription-Token, error classes),
  result normalization with redaction/truncation, SSRF guard, HTML→text fetch
  (10s, 512KB, text/html|text/plain only)
- modules/websearch: 10-min PG cache (region websearch), cross-process
  per-minute limit, web.fetch allowlist = URLs from the caller's latest
  web.search; POST /websearch/test
- settings.agent.webSearch (key write-only → secret_ref); settings card
  under Agent; web.* offered to the model only when enabled, plus a system
  prompt rule; outputs marked untrusted
- capability count 27 → 29; e2e E13; runbook docs/runbook/websearch.md

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-29 00:14:08 +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 Fable 5.1 fa261eb6cd chore: ignore .local/ (test-only credentials)
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 23:23:20 +08:00
rootandClaude Fable 5.1 c62693079f docs(plan): camofox shipped as default compose service
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 23:22:05 +08:00
rootandClaude Opus 5.5 f05d6195a3 fix(build): compiled binary missing @modelcontextprotocol/sdk modules
shadcn (web devDependency) pulled a second peer copy of
@modelcontextprotocol/[email protected].1 (zod@3), so the server's zod@4 copy
lived in the npm cache as sdk/1.30.1_1. `deno compile --exclude-unused-npm`
embeds that _1 folder but the pruned snapshot resolves it as sdk/1.30.1,
so every SDK import failed at startup with ERR_MODULE_NOT_FOUND.

- drop the shadcn CLI dependency; vendor its tailwind.css (the only
  runtime use) as apps/web/app/shadcn-tailwind.css, removing the duplicate
- scripts/check-compiled-binary.sh: run the compiled binary with an empty
  env and `--role preflight`; it must load the full module graph and exit 2
  on config validation. Wired into `deno task ci` (check:binary) and the
  Dockerfile build stage.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 23:10:08 +08:00
rootandClaude Fable 5.1 266d66ddbb docs(plan): lessons learned from multi-agent parallel development
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 22:47:20 +08:00
root cccedb9d7d Merge branch 'c2a' 2026-09-28 22:26:58 +08:00
rootandClaude Opus 5.5 7862573944 feat(cache): TMDB image proxy, posters, cache admin API and settings cards
GET /img/tmdb/{size}/{file} ported from MP ImageHelper / fetch_image
(sanitized cache path under DATA_DIR/cache/images, GLOBAL_IMAGE_CACHE
read switch, md5 ETag / 304, day-based cleanup job). Cache module adds
cache.sweep / cache.images_clean jobs and /cache/tmdb, /cache/images
list and clear endpoints. Library list/detail and the search palette
show posters with the neutral placeholder as fallback; metadata settings
get TMDB cache and image cache cards. Runbook docs/runbook/cache.md and
porting notes docs/porting/moviepilot-tmdb-cache.md.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 22:26:29 +08:00
rootandClaude Opus 5.5 946fca5e55 feat(cache): PG kv_cache ported from MP TTLCache, TMDB request and recognize caches
Migration 0042_cache adds UNLOGGED kv_cache. PgCache mirrors MP
runtime/cache.py (region, per-key TTL, TLRU maxsize eviction, @cached as
cachedCall); only the storage moves from memory to PG. TMDB request
responses (MP TMDb.request @cached) and recognize results (MP TmdbCache:
same key, {id:0} negative cache with EMPTY_RESULT_CACHE_TTL, type-conflict
checks, backfill after manual assignment) live in kv_cache, shared by all
processes. Settings cache.tmdbTtlHours / tmdbMaxSize / emptyResultTtlSec
map to META_CACHE_EXPIRE / CONF.tmdb / EMPTY_RESULT_CACHE_TTL. TMDB
details now also store posterPath / backdropPath.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 22:26:29 +08:00
rootandClaude Fable 5.1 c3ba591b02 chore(deploy): keep docker env in .env.deploy, restore .env for development
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 22:08:24 +08:00
rootandClaude Opus 5.5 d2b5095e97 docs(plan): v0.2.1 independent check report
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 22:07:53 +08:00
rootandClaude Opus 5.5 c1c8dc81a9 docs: requirements coverage matrix and next-phase plan
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:52:25 +08:00
rootandClaude Opus 5.5 61f585a707 feat(sites): cross-process site rate limit via PG token bucket
Migration 0041_rate_limit adds rate_limit_bucket. PgTokenBucket (@umt/db)
refills and takes tokens under SELECT ... FOR UPDATE on the DB clock and
polls with jitter within a wait budget; penalize() carries Retry-After.
Site adapters get createSharedSiteRateLimiter (key site:<id>), shared by
api / worker / worker:agent, with an in-process serial queue in front and
a local fallback when the bucket is unavailable.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:51:32 +08:00
rootandClaude Opus 5.5 57346d438c fix(web): replayed tool call after resume reuses its card (duplicate React key)
A suspended agent.propose_options / approval call is replayed on resume with the
same toolCallId; foldEvents pushed a second item with the same key, which also
inflated the tool call meter. Also use 「」 in the report footnote.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:41:44 +08:00
rootandClaude Fable 5.1 a99c77ce4e docs(plan): TMDB response cache in PG, image proxy with disk cache (no Web Cache API)
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 21:40:53 +08:00
rootandClaude Fable 5.1 5169529562 docs(plan): cache and rate-limit decisions (PG token bucket, unlogged cache tables, no Redis)
Co-Authored-By: Claude Fable 5.1 <[email protected]>
2026-09-28 21:39:25 +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 0fccfe257d fix(web): settings side nav lists MCP, capabilities and passkey anchors
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:34:51 +08:00
rootandClaude Opus 5.5 410d4eef6f chore: sort mcp export and gen:types output, sort deno.json imports
HEAD failed biome organizeImports on packages/contracts/src/api.ts; types.ts
now matches deno task gen:types on a fully migrated dev database.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:09:25 +08:00
rootandClaude Opus 5.5 cd70a69508 docs(runbook): automation-first agent interaction, plan/options/approval events and E12
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:07:01 +08:00
rootandClaude Opus 5.5 9649854ab3 feat(web): live step list, option choice card and inline approval card; drop run mode switch
- session page shows the agent.plan TodoList (drawer: one-line progress); options and approval
  cards sit at their step; composer has a "先看方案再执行" toggle instead of the mode switch
- settings expose option timeout, timeout action and approval validity; quick goals follow glossary

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:07:01 +08:00
rootandClaude Opus 5.5 cf2c4ff0e6 feat(agent): automation-first sessions with step plan, option choice and inline approval
- agent.plan emits plan_updated; system prompt asks for a plan first and updates while executing
- agent.propose_options registers a decision and suspends the run (waitForFact, no worker slot);
  the answer or timeout (recommended/stop) resumes it as the tool result
- console-session runs get inline approvals for plan-bound capabilities: approve replays the call
  with the approval, reject fails it with the reason and records it on the decision
- proposeFirst blocks execute capabilities until an option is chosen; E12 covers all paths

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:07:01 +08:00
rootandClaude Opus 5.5 74c2ed45aa feat(contracts,db): agent.plan / agent.propose_options, interaction events, proposeFirst replaces run mode
- agent_run_event gains plan_updated, options_proposed/resolved, approval_requested/resolved (migration 0040)
- AgentRunStartInput.proposeFirst; mode deprecated (plan_only maps to proposeFirst); mode dropped from AgentRun API
- POST /agent/runs/{id}/choice; agent.optionsTimeoutSec/optionsTimeoutAction/approvalTimeoutSec settings

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 21:07:01 +08:00
rootandClaude Opus 5.5 d931e75310 docs(runbook): MCP server access and security boundary
Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 20:28:41 +08:00
rootandClaude Opus 5.5 b3cc5a8313 feat(web): tools and MCP section in Agent settings, tool call IO drawer
Settings > Agent: MCP server card (endpoint, enable switch, allowed API keys,
live sessions, recent calls, copyable client config without secrets) and the
capability list (scope, risk, idempotency/replay, authorization state, 7-day
calls and failure rate, last call, schema sheet). Agent session tool cards get
a drawer with input/output schema and the redacted input/output of the call.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 20:28:41 +08:00
rootandClaude Opus 5.5 653100ceb4 feat(mcp): Streamable HTTP MCP server on the capability registry
POST/GET/DELETE /api/v1/mcp with API Key auth; tools are the capabilities
visible to the key's scope (download__submit naming), every call goes through
registry.invoke (authorize, audit, idempotency key, ledger) with source=mcp.
Resources: settings summary, pending decisions, subscriptions overview;
prompts: triage_pending, acquire_title, diagnose_task. Sessions bound to the
API key, idle reaping, per-key rate limits, enable switch and key allowlist.
Also GET /capabilities/stats (from audit_event) and /capabilities/tools.
Tests connect with the official SDK client over Streamable HTTP.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-09-28 20:28:34 +08:00