18 Commits

Author SHA1 Message Date
Ryan Chen 609ee5aa7e Record Immich 3.3.1 and make the conductor sole owner of tickets.md
CI / backend (push) Successful in 43s
CI / frontend (push) Successful in 30s
CI / backend (pull_request) Successful in 42s
CI / frontend (pull_request) Successful in 29s
Production Immich is reachable at immich.ryanjchen.com and runs 3.3.1,
so the filter API style is the default and the public URL is known.
Parallel ticket PRs editing one status file would conflict, so only the
conductor updates tickets.md, once per wave.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 09:20:35 -04:00
Ryan Chen 29200e49db Add Immich MCP tickets and execution order
CI / backend (push) Successful in 43s
CI / frontend (push) Successful in 31s
CI / backend (pull_request) Successful in 41s
CI / frontend (pull_request) Successful in 29s
Breaks the merged design into eight tickets (IMM-01..08) with files,
steps, tests and acceptance criteria, the wave order with what can run
in parallel, and the Herdr conductor/worker rules used to build them.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 09:08:15 -04:00
Ryan Chen 200cd73924 Add Immich MCP technical design
CI / backend (push) Successful in 42s
CI / frontend (push) Successful in 9m34s
CI / backend (pull_request) Successful in 42s
CI / frontend (pull_request) Successful in 9m34s
Design for an in-repo stdio MCP server that lets Simba (and Claude Code)
find existing albums, summarize trips by date clusters, and create or
fill albums behind a write flag and a confirmed-count check.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 08:40:58 -04:00
Ryan Chen b02bf0a6df Sync Obsidian both ways so notes Simba writes reach your devices
CI / backend (pull_request) Successful in 42s
CI / frontend (pull_request) Failing after 6s
CI / backend (push) Successful in 41s
CI / frontend (push) Failing after 7s
The container ran Obsidian Sync in pull-only mode, so notes, tasks and journal
entries the agent wrote stayed in its local copy. OBSIDIAN_SYNC_MODE now
defaults to bidirectional (merge on conflict); pull-only and mirror-remote
remain available.

Blue and green share the vault volume, and two sync clients writing one vault
and its sync state would race during a deploy. Setup and sync now run under a
flock on a file in the vault, so a new container starts syncing only after the
old container's sync process exits. Clearing Obsidian's stale sync lock is safe
inside that lock. Setup runs in the background, so startup no longer waits for
Obsidian login.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 08:10:48 -04:00
Ryan Chen 5809ad3adb Add durable chat runs, shared agent runtime, reliable jobs, history tools and citations
CI / backend (pull_request) Successful in 42s
CI / frontend (pull_request) Failing after 7s
CI / backend (push) Successful in 41s
CI / frontend (push) Failing after 7s
Shared runtime
- One agent entry point (runtime.invoke_agent_for_user / prepare_agent_turn) for
  web, email, iMessage, WhatsApp, scheduled and cron, so identity, permissions,
  memories, summaries and metrics match across channels.
- Serialize turns per checkpoint thread; /query validates input and returns 409
  while a durable run owns the conversation.

Reconnectable chat (durable runs)
- POST /runs creates an idempotent, owner-scoped run; GET .../events replays an
  ordered SSE log by cursor, so generation survives refresh and device switches.
  Stop cancels; disconnect only detaches. /stream-query is a compatibility adapter.
- Coalesce streamed deltas (150 ms) and wake same-process subscribers instead of
  polling; keep buffered text when a run is stopped.
- Drain runs on shutdown; serve via Hypercorn with a shutdown timeout covering
  the drain (and no longer in debug mode). CHAT_RUN_* settings are forwarded.

Scheduled messages and cron
- Atomic delivery claims, attempt history and explicit retries; ambiguous sends
  become "uncertain" and are never resent automatically.
- Cron triggers need a run_id/Idempotency-Key; legacy jobs need an explicit owner.
- Cancelling never rewrites a started delivery; series stop via tombstones.
  Legacy sent rows are not forked into new series. Malformed bodies return 400.

History, integrations and retrieval
- Message-content search, pinned conversations, JSON/Markdown export and a
  memory editor.
- Integration status panel and YNAB data-freshness provenance.
- Date-filtered document/note search with stable citations and a source viewer.

Migration 20 is generated by aerich.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 23:11:04 -04:00
Ryan Chen de88679f25 Validate production config, add CI, and build from locked dependencies
CI / backend (pull_request) Successful in 1m23s
CI / frontend (pull_request) Failing after 37s
CI / backend (push) Successful in 36s
CI / frontend (push) Failing after 7s
- Fail startup on unsafe or malformed production settings (scripts/validate_config.py),
  run it before migrations, and stop on a failed `aerich upgrade`.
- Forward the context-budget, timeout and metrics settings through Compose; accept
  the historical TAVILIY_API_KEY spelling.
- Build the image from uv.lock and yarn.lock in a multi-stage Dockerfile.
- Add Gitea Actions CI: Ruff, unit tests, context evaluation, frontend tests,
  typecheck and build.
- Update README and deployment/development docs to match the current stack.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 23:09:22 -04:00
Ryan Chen 4645920c64 Route tools and retain structured conversation context with durable evidence 2026-10-12 20:19:12 -04:00
Ryan Chen 1c9405cfdf Budget conversation context while preserving current turns and tool exchanges 2026-10-12 19:45:51 -04:00
Ryan Chen 10f3f1bcab Merge origin/main into feat/ynab-mirror
Resolve the Makefile .PHONY conflict (desktop + ynab targets) and
regenerate the YNAB migration with aerich as 17 on top of main's
16_add_oidc_states, replacing the colliding 16_add_ynab_mirror.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 18:06:47 -04:00
Ryan Chen abf9415696 Add Cowork UI (phase 2) plan and drop quick-ask window
Defines the shared contract (SSE v2 tool events, artifact kinds and
storage, artifacts API, uploads), five parallel tasks with file
ownership, merge order and verification, and removes the quick-ask
window from the desktop plan.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 17:47:43 -04:00
Ryan Chen eea96d9010 Add desktop client exploration doc
Compares PWA, Tauri v2 and Electron shells, recommends a Tauri shell
loading the deployed origin, and lays out backend changes (structured
SSE events, artifacts table, uploads, notifications, local tool bridge,
access-token auth) with a phased plan.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 17:40:49 -04:00
Ryan Chen 8f02318542 Mirror YNAB budget into Postgres with delta sync and change log
YNAB has no event feed, so keep a local copy in ynab_* tables synced via
server_knowledge delta requests: full download on first run, deltas every
YNAB_SYNC_INTERVAL seconds and on demand before each tool call, full
resync daily. Each sync diffs against the mirror and appends real edits
(transactions, assignments, renames) to ynab_changes, the edit history
YNAB doesn't keep.

Agent YNAB tools now answer from the mirror: budget summary (spending
computed from transactions, since YNAB's month activity nets in credit
card payment categories), account balances, transaction search with
splits expanded, spending reports by category/payee/account/month,
category budget, and recent changes. Adds an admin-only read-only
ynab_sql tool guarded by a sqlglot allowlist (scope-resolved tables, no
dangerous functions), a READ ONLY transaction, a timeout, and a row cap.

Replaces utils/ynab_service.py and drops the unused ynab SDK.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 17:32:38 -04:00
Ryan Chen d8d01a5fff Zero-downtime blue-green deploys behind Caddy health checks
Split the raggr service into raggr-blue (8080) and raggr-green (8081)
behind compose profiles, add a /api/health endpoint and container
healthcheck, and add scripts/deploy.sh to build the inactive color,
wait for it to become healthy, then gracefully stop the old one.
Makefile targets resolve the active color automatically.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-09-10 01:41:37 -04:00
Ryan Chen 6ae36b51a0 ynab update 2026-01-31 22:47:43 -05:00
Ryan Chen 713a058c4f Adding roadmap 2026-01-31 17:28:53 -05:00
Ryan Chen ad39904dda reorganization 2026-01-31 17:13:27 -05:00
Ryan Chen 1fd2e860b2 nani 2026-01-31 16:47:57 -05:00
Ryan Chen 7cfad5baba Adding mkdocs and privileged tools 2026-01-31 16:20:35 -05:00