llama-server only reuses a byte-identical prompt prefix. Several things
changed the prefix every turn, forcing most of the prompt to be re-prefilled:
- Memories: rank still chooses which memories fit the budget, but the chosen
ones are emitted in stored order, so the block is identical across queries.
- History trimming: the conversation-evidence summary is only included when
turns were actually dropped (it duplicated retained turns otherwise). On
overflow, old turns are dropped at stable cut points down to ~65% of the
budget, so the retained prefix holds for several turns instead of shifting
by one turn every request.
- Tools: TOOL_ROUTING_MODE=static (default) binds the full role-appropriate
tool set in registration order with no discovery tool; admin gating is
unchanged. TOOL_ROUTING_MODE=dynamic keeps the keyword router.
- Observability: per-call model_calls in the stream_request log now include
cached_prompt_tokens when the provider reports cache reads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 30s time-to-first-token guard measured from the turn's first model call
to the first answer token, so thinking, tool execution and the second call's
prefill all counted against it. Long reasoning on a long conversation was
cancelled mid-generation (llama-server: stop: cancel task at n_gen=1287).
- stream_query: the timeout now fires only when an in-flight model call goes
silent for TTFT_TIMEOUT_SECONDS. Any chunk (thinking or empty) resets the
clock, each model call starts a fresh one, and time spent in tools between
calls doesn't count.
- Trim replayed history (blueprints/conversation/context.py): tool results
and user messages from earlier turns are capped at 4000 chars (attachments
included) and their images replaced by a placeholder. The current turn is
untouched. The window no longer starts with orphaned ToolMessages.
- Log per-call prefill time, duration and prompt/output tokens
(stream_usage=True) plus max_prompt_tokens in the stream_request line.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Convert mealie_get_recipe, mealie_meal_plans, mealie_todays_meals,
mealie_shopping_lists, obsidian_read_note, journal_get_today,
obsidian_search_notes and simba_search to content_and_artifact. The text
the LLM sees is unchanged; the artifact is a JSON-safe ToolArtifact dict
per the Cowork UI contract (mealie.recipe, mealie.meal_plan,
mealie.shopping_lists, obsidian.note, search.documents). Error, empty and
not-configured paths return no artifact.
simba_search's artifact changes from a list of Documents (unused) to a
search.documents dict.
New optional env vars: MEALIE_GROUP_SLUG (default "home") for Mealie v2
recipe links, OBSIDIAN_VAULT_NAME for obsidian://open links.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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>
Load an allowlisted subset of bookstack-mcp tools (search/read plus
create/update books, chapters, pages and comments; no deletes) over stdio
and add them to the admin agent alongside Homebox. Extract the shared
allowlist + is_admin wrapping into utils/mcp_tools.py.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Load an allowlisted subset of homebox-mcp tools (read + safe writes, no
deletes) over stdio and compile them into a separate admin agent, so
non-admins never see them. Each tool also re-checks is_admin per call.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Multi-tenant system to fetch and store restaurant lists from Beli accounts.
Each user can connect up to 3 Beli profiles; a nightly Cronicle job (1–5 AM,
randomised) refreshes the 3 stalest profiles using the reverse-engineered Beli
API (mirrored at git.torrtle.co/ryan/beli-api). Rankings (been) and bookmarks
(want-to-try) are stored in Postgres and queryable via a new LangChain agent
tool. Beli passwords and refresh tokens are stored Fernet-encrypted.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Switches from curl (shell not available in Cronicle container) to a Node.js
inline script using the built-in http module. Tries raggr-green first, falls
back to raggr-blue automatically — no manual update needed on color swap.
Delete the existing broken Cronicle event before redeploying.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Switch shellplug → urlplug to avoid spawn errors in shell-less containers
- Explicitly set timing to all 60 minutes (Cronicle {} defaults to hourly :00)
- Restore RAGGR_INTERNAL_URL env var for the HTTP plugin target URL
Delete the existing broken Cronicle event before redeploying so it gets recreated.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Shell script falls back to the other color if the primary is down,
so the trigger works regardless of which container is currently active.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
On startup raggr calls the Cronicle API (CRONICLE_URL + CRONICLE_API_KEY)
and creates a 'Scheduled Messages Trigger' shell event if it doesn't exist,
replacing the need to configure it manually in the Cronicle UI.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces per-request history reconstruction with a persistent AsyncSqliteSaver
checkpointer keyed by conversation ID, so agent state (tool calls, intermediate
steps) carries forward between turns without manual message list building.
Also includes the missing migration for the CronJob model added in bb4619a.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a new cron blueprint with UUID-keyed jobs that Cronicle triggers via
HTTP POST with a shared secret header. Also adds scripts/migrate-new.sh so
migrations can be generated locally against a Docker postgres when no prod
container is running.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces the gws CLI subprocess approach with google-api-python-client.
Adds get_calendar_events, create_calendar_event, gmail_search_emails, and
gmail_get_email as LangChain tools; all admin-gated and auto-enabled when
token.json is present. Includes uv-runnable scripts/google_auth.py for
one-time OAuth2 setup.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Tracks end-to-end response time and request counts (by stream mode and
result) via prometheus_client. Exposes GET /metrics behind a bearer
token (METRICS_TOKEN env var) so Prometheus can scrape it without
exposing raw data publicly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When llama-server is queue-backed it accepts the connection but never
produces tokens; after ~60s Caddy drops the connection and the frontend
clears loading state with no response and no error shown to the user.
Backend: replace the async-for event loop with an explicit __anext__()
iterator wrapped in asyncio.wait_for(). A 30s timeout (TTFT_TIMEOUT_SECONDS)
applies only in the window between on_chat_model_start and the first content
token — tool calls are excluded. On timeout, emits a stalled error event and
logs it rather than waiting for the proxy to kill the connection.
Frontend: separates stall errors from generic errors; shows a "Simba didn't
respond in time. Retry" pill that re-sends the original query.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds simbakit/ as a subfolder service that polls PetKit smart devices
(litter box, fountain, feeder) and exposes pet weight, litter activity,
and device health data via three new LangChain tools.
Also fixes stale alerts bug where cleared conditions (e.g. waste bin
full) persisted because the query used max alert time instead of the
device's last_seen_at timestamp.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Validates sb-signing-secret header against SENDBLUE_WEBHOOK_SECRET env var.
Can be disabled with SENDBLUE_SIGNATURE_VALIDATION=false for development.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- New imessage blueprint: webhook receives inbound iMessages, runs through
LangChain agent, replies via SendBlue REST API
- Admin-only: only users with lldap_admin group can use iMessage channel
- Admin endpoints to link/unlink imessage_number on user accounts
- Add imessage_number field to User model (needs aerich migration)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a get_calendar_events agent tool that shells out to `gws calendar +agenda`
for admin users. Controlled by GOOGLE_CALENDAR_ENABLED env var, with OAuth
credentials mounted from credentials.json.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace OBSIDIAN_AUTH_TOKEN with OBSIDIAN_EMAIL/OBSIDIAN_PASSWORD and run
the full ob login → sync-setup → sync sequence on container startup so
fresh containers authenticate properly instead of failing with
"Run 'ob sync-setup' first".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds EMBEDDING_SERVER_URL and EMBEDDING_MODEL_NAME env vars, mirroring
the existing LLAMA_SERVER_URL pattern for LLM configuration.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Consolidate onto PostgreSQL by using pgvector instead of a separate
ChromaDB instance. This removes a Docker volume, a large dependency,
and simplifies the stack without meaningful performance impact at
our document scale.
- Swap langchain-chroma for langchain-postgres (PGVector)
- Use pgvector/pgvector:pg16 Docker image with init script
- Lazy-initialize vector store to avoid eager DB connections
- Add SQL helpers for stats/delete/list (replacing _collection access)
- Remove legacy main.py, chunker, petmd scraper, and /api/query endpoint
Re-index required after deploy (POST /api/rag/index + /index-obsidian).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Users can now receive a unique email address (ask+<token>@domain) and
interact with Simba by sending emails. Inbound emails hit a Mailgun
webhook, are authenticated via HMAC token lookup, processed through the
LangChain agent, and replied to via the Mailgun API.
- Extract shared SIMBA_SYSTEM_PROMPT to blueprints/conversation/prompts.py
- Add email_enabled and email_hmac_token fields to User model
- Create blueprints/email with webhook, signature validation, rate limiting
- Add admin endpoints to enable/disable email per user
- Update AdminPanel with Email column, toggle, and copy-address button
- Add Mailgun env vars to .env.example
- Include database migration for new fields
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add FERNET_KEY to .env.example with generation instructions
- Register email.models in aerich_config.py and app.py
- Register email blueprint in app.py
- Create database migration for email_accounts, email_sync_status, emails
- Migration includes proper foreign keys and indexes
- Update llm.py to use OpenAI client with custom base_url for llama-server
- Update agents.py to use ChatOpenAI instead of ChatOllama
- Remove unused ollama imports from main.py, chunker.py, query.py
- Add LLAMA_SERVER_URL and LLAMA_MODEL_NAME env vars
- Remove ollama and langchain-ollama dependencies
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add DATABASE_PATH environment variable support in app.py and aerich_config.py
- DATABASE_PATH: For simple relative/absolute paths (default: database/raggr.db)
- DATABASE_URL: For full connection strings (overrides DATABASE_PATH if set)
- Create .env.example with all configuration options documented
- Maintains backward compatibility with default database location
Usage:
# Use default path
python app.py
# Use custom path for development
DATABASE_PATH=dev.db python app.py
# Use full connection string
DATABASE_URL=sqlite://custom/path.db python app.py
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Ryan Chen