Everything below the run loop already existed. This is the thing that runs a model. The build tag is gone, deliberately. internal/agent's doc comment promised two separations — cmd/pansy not importing the package, and the tool wiring behind //go:build majordomo — and both have been rewritten rather than left as a stale aspiration. A tag that keeps the agent out of the binary only earns its keep if you would ever ship a build without the agent, and the agent is the point; keeping it meant an untagged CI that never compiled the code that matters. majordomo is a real dependency now, resolved from the Gitea instance as a pseudo-version with no replace directive, so the Docker build (which has no sibling checkout) resolves it the same way this machine does. It is stdlib-first and pure Go, so CGO_ENABLED=0 and the single static binary survive. A TURN IS ONE CHANGE SET. That is the whole reason acting without a confirmation prompt is defensible: "empty the garlic bed and plant cucumbers" is one object edit and a dozen planting inserts, and it has to undo as one action rather than thirteen. The scope is opened even for a turn that turns out to be a question, because a change set with no revisions is never written — so asking costs nothing and history isn't littered with empty entries. The model spec goes to majordomo.Parse verbatim. That grammar, including comma-separated failover chains, is majordomo's; re-implementing any of it here would only mean two places to update when it grows. The key needs a bridge though: majordomo's ollama-cloud preset reads OLLAMA_API_KEY while pansy (like gadfly) is configured with OLLAMA_CLOUD_API_KEY, so the provider is registered explicitly on a private registry rather than depending on ambient environment. Runs are bounded by a step cap, a timeout and majordomo's loop guards. This is loop safety, not cost control — pansy is a personal tool and spend caps are explicitly not a v2 concern. A capped run does NOT fail: it kept whatever it managed to do, that work is recorded and undoable, and the reply says it stopped early rather than going silent. The chat endpoint streams. A turn that clears a bed and replants it makes a dozen tool calls over tens of seconds, and without streaming that is a long silence followed by everything at once — which reads as a hang, and defeats a design that rests on watching the canvas change as it happens. Conversations persist per (user, garden). Client-held history would be lost on a refresh, which is exactly when someone reloads to check whether the agent's change landed. Only the user/assistant TEXT is stored, not the model's full transcript: continuity needs what was said and what came back, and replaying a stored tool call would replay a decision made against a garden that has since moved on. It also keeps majordomo's message shape out of the schema. An instance with no key starts, serves the app, and doesn't advertise the agent — the routes aren't registered at all, the same shape as OIDC 404ing when unconfigured. A configured-but-unresolvable model logs and disables the assistant rather than refusing to boot: a garden planner that won't start because of a chat feature is worse than one without chat. Tool refusals reach the model as tool results it can explain, not 500s. The ACL story only works if it can narrate the refusal. Closes #56 Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]> Claude-Session: https://claude.ai/code/session_01H3zbym8Doka2d7D48maSgZ
17 KiB
Pansy — Design
Pansy is a self-hostable garden planning web app. A landscaping-style 2D editor: drag garden objects (raised beds, grow bags, containers, in-ground plots, trees, paths) onto a field at real-world scale; click into a plantable object to place freeform "plops" of plants — a few plops of garlic in one corner, herbs in another, beans along a side — and zoom out to still see what's planted where. React frontend, Go backend, clean JSON API, single static binary. Works on desktop and is mobile-usable (touch drag, pinch zoom).
Work is tracked in Gitea issues; the tracking epic links every piece in dependency order.
Decisions
- Placement model: freeform plops (not a square-foot grid), scaled by real plant spacing. Grid snapping may come later as a toggle.
- Stack: Go 1.26.x backend, module
gitea.stevedudenhoeffer.com/steve/pansy; React + TypeScript + Vite + Tailwind frontend, production build embedded viaembed.FS→ one static binary (CGO_ENABLED=0). - Users: multi-user with ownership. Users own gardens; a garden can be shared with other users as viewer (read) or editor (edit content). Owner additionally shares/deletes.
- Auth: OIDC-first (Authentik is the primary IdP), local argon2id passwords as an optional fallback.
- Agentic future: integration with majordomo/executus via typed Go tools (
llm.DefineTool[Args]) wrapping the same service layer the REST API uses — not MCP/OpenAPI.
Domain model
SQLite, centimeters everywhere (display-side imperial conversion only), version INTEGER on every user-mutable row for conflict detection.
- users — email (unique), display_name, nullable argon2id
password_hash, nullableoidc_issuer+oidc_subject(unique pair),is_admin. First registered user becomes admin.PANSY_REGISTRATION(open/closed) gates local signup. A user can be local-password, OIDC, or both (linked by email on first OIDC login). - sessions — sha256-hashed random 32-byte token, user_id, 30-day sliding expiry. HttpOnly cookie holds the raw token. Sessions are pansy-local regardless of auth method.
- gardens — owner_id, name, width_cm, height_cm, unit_pref (
metric|imperial), notes. - garden_shares — (garden_id, user_id) unique, role
viewer|editor, created_by. Owner is implicit viagardens.owner_id, never a share row. - garden_objects — one polymorphic table:
kind(bed|grow_bag|container|in_ground|tree|path|structure), name,shape(rect|circle;polygon+pointsJSON reserved in schema, not in the v1 editor), centerx_cm,y_cmin garden space,width_cm,height_cm(circle: width=diameter),rotation_deg,z_index,plantablebool, nullablecoloroverride,propsJSON for kind-specific extras (bed height, tree species…), notes. - seed_lots — one purchase: vendor, source URL, sku, lot code, purchase date, packed-for year, quantity + unit, cost, germination %.
owner_idis on the LOT, not inherited from the plant, so you can record buying generic built-in Garlic from Johnny's and keep it private.plantings.seed_lot_id(ON DELETE SET NULL) attributes a plop to a purchase;remainingis derived (quantity − summed effective count of active plantings), never stored — a decremented column drifts the moment a planting is edited behind its back and can't be audited afterwards. Lots are never shared along with a garden. - plants — catalog, plus
source_url/vendorprovenance for the variety itself.owner_id NULL= built-in (~30 seeded via migration, read-only, clone to customize). name, category (vegetable|herb|flower|fruit|tree_shrub|cover),spacing_cm(mature spacing),colorhex,icon(emoji string — zero assets in v1), nullabledays_to_maturity, notes. Users see built-ins + their own. - agent_conversations / agent_messages — the assistant's chat, one thread per (user, garden). Only the user/assistant TEXT of each turn is stored, not the model's full transcript: continuity needs what was said and what came back, and replaying a stored tool call would replay a decision made against a garden that has since moved on.
agent_messages.change_set_idis the handle the chat panel uses to offer Undo on the turn that caused it. - journal_entries — the grow log:
garden_id(NOT NULL), nullableobject_id/planting_idto narrow the target, author, body,observed_at.noteson a garden/object/plant says what the thing IS and a new note overwrites the old; a journal entry says what HAPPENED and when, and entries accumulate.garden_idis set even for a bed- or plop-level entry so permission checks reuserequireGardenRoleunchanged rather than inventing a second ACL path.observed_atis distinct fromcreated_at— you write up Saturday's observations on Sunday. Deliberately not in the revision history: entries are already append-shaped and individually versioned. - change_sets / revisions — the undo substrate. A change set groups the row-level revisions one logical operation produced (
sourceui/agent/api,summary, nullableagent_run_id, nullablereverts_id); a revision is(entity_type, entity_id, op, before, after)with full JSON row snapshots. The unit of undo is the operation, not the row. Revert is itself a change set, so an undo can be undone; it is version-guarded per entity and reports conflicts rather than clobbering a later edit. Garden deletion is deliberately not revertible (the cascade is invisible to the service, and would take the history with it). - plantings ("plops") — object_id, plant_id, center
x_cm,y_cmin the parent object's local frame (moving/rotating a bed moves its plants for free),radius_cm(a plop is a circular patch), nullablecount(NULL = derived:max(1, round(π·r² / spacing_cm²))), nullable label,planted_at/removed_atdates. "Clear bed" setsremoved_at; v1 shows rows whereremoved_at IS NULL. Year-over-year history and a future plan/scenario layer build on these dates without schema surgery.
Geometry
- Garden space: origin top-left, x→right, y→down (screen-aligned), cm.
- Objects: positioned by center point +
rotation_degclockwise about center → SVGtranslate(x y) rotate(deg)one-liners. - Object local frame: origin at object center, axes aligned to the unrotated object. Plops live in this frame.
- Exactly two nesting levels: garden → object → plop. No recursive containers.
- API and DB are metric-only; cm ↔ ft/in conversion is purely a frontend display concern.
Editor / rendering
- Plain SVG in React. Tens of objects + low-hundreds of plops is far below SVG's ceiling; native DOM hit-testing, Tailwind styling, crisp text at any zoom. No Konva/canvas.
- One gesture library:
@use-gesture/reactfor unified drag / wheel-zoom / pinch on desktop + touch. Everything else is hand-rolled pointer math. - Viewport = a single
<g transform="translate(tx,ty) scale(s)">; state{tx, ty, scale}where scale = px per cm. Wheel/pinch zooms to cursor; drag on empty space pans; drag on an element moves it. - Field view and bed interior are the same canvas. "Click into a bed" animates the viewport to fit the object and sets
focusedObjectId(mirrored to?focus=URL param for deep links). Focused mode dims siblings and enables plop placement; Escape/tap-out zooms back. - Semantic zoom, three bands by scale (thresholds tuned by feel):
- zoomed out (< ~0.75 px/cm): plops render as flat color patches; plantable objects show name + dominant-plant color — "what's planted where" at a glance;
- mid: plops as colored circles with the plant's emoji icon;
- zoomed in (> ~3 px/cm): icon + plant name + count per plop.
API
REST under /api/v1, JSON, cookie-session auth (HttpOnly, SameSite=Lax, Secure under TLS). Gin with slog-gin + Recovery; embedded SPA fallback for non-/api routes.
POST /auth/register | /auth/login | /auth/logout GET /auth/me GET /auth/providers
GET /auth/oidc/login GET /auth/oidc/callback (authorization-code + PKCE)
GET,POST /gardens GET,PATCH,DELETE /gardens/:id
GET /gardens/:id/full ← one-shot editor load: garden + objects + plantings + referenced plants
GET /gardens/:id/full?year=YYYY ← season view: plops whose time in the ground overlapped that year
GET /gardens/:id/years ← years this garden has planting data for
GET /gardens/:id/history ← change sets, newest first (paginated)
POST /change-sets/:id/revert ← undo an operation; 201, or 409 + the conflicts it skipped
POST /gardens/:id/copy ← deep-copy a garden you own (objects + active plops; not shares/link)
POST /gardens/:id/objects PATCH,DELETE /objects/:id
POST /objects/:id/plantings PATCH,DELETE /plantings/:id
GET,POST /plants PATCH,DELETE /plants/:id (own plants only)
GET,POST /seed-lots GET,PATCH,DELETE /seed-lots/:id (own lots only; private)
GET,POST /gardens/:id/journal PATCH,DELETE /journal/:id (editor writes; author edits own)
GET /gardens/:id/journal/counts ← entries per object, for the "has notes" indicator
POST /agent/chat ← SSE: step events, then the finished turn (editor only)
GET,DELETE /gardens/:id/agent/history (the actor's own thread)
GET,POST /gardens/:id/shares PATCH,DELETE /gardens/:id/shares/:userId (invite by email)
Sync: plain REST + optimistic UI + last-write-wins with a version guard. Every PATCH/DELETE carries the row's version; the server increments on write and returns 409 + the current row on mismatch; the client rolls back and refetches. No websockets/CRDT — the right cost for household-scale co-editing. Drags PATCH once on drop, not per frame.
Backend layout
Storage: modernc.org/sqlite (pure Go) + hand-written SQL via database/sql. WAL mode + busy_timeout=5000 (open pattern per executus contrib/store/sqlite.go). Migrations are numbered .sql files in an embed.FS with a tiny runner and a schema_migrations(version) table, run at startup; seed plants live in a migration.
cmd/pansy/main.go config load → store.Open → migrate → service → api → ListenAndServe
internal/config/ PANSY_PORT, PANSY_DB, PANSY_BASE_URL, PANSY_REGISTRATION,
PANSY_OIDC_* (issuer/client_id/secret/button label), PANSY_LOCAL_AUTH,
PANSY_AGENT_MODEL / OLLAMA_CLOUD_API_KEY / PANSY_AGENT_ENABLED,
trusted proxies
internal/store/ sqlite.go (open/pragmas), migrate.go, migrations/*.sql (embedded),
queries per entity (users.go, gardens.go, objects.go, plantings.go,
plants.go, shares.go)
internal/domain/ plain structs + sentinel errors (ErrNotFound, ErrForbidden,
ErrVersionConflict)
internal/service/ ★ THE SEAM. All permission checks + business ops. auth.go, gardens.go,
objects.go, plantings.go, plants.go, shares.go, ops.go (FillRegion etc.),
revisions.go (change-set scoping, recording, revert)
internal/api/ thin gin handlers (decode → service → encode), session middleware,
ginserver setup, routes.go, oidc.go, spa.go (embed fallback)
internal/webdist/ dist.go: //go:embed all:dist (web build copied in by Makefile)
internal/agent/ tools.go (llm.DefineTool wrappers over the SAME service methods),
runtime.go (model resolution, run loop, one change set per turn)
web/ Vite app
Makefile (cd web && npm run build) → copy dist → CGO_ENABLED=0 go build
Every mutation is recorded. Service mutations call record(...), which joins the change set on the context if one is open and otherwise opens a single-op one for that mutation alone. That auto-scope is what keeps undo shallow: REST handlers need no changes at all, and only the agent wraps a whole turn (WithChangeSet) so a multi-step turn undoes as one unit. A multi-row operation passes all its changes in one record call for the same reason.
The service seam is the point. Every operation is a method on *service.Service taking (ctx, actorUserID, args); all ACL checks live there, not in handlers. REST handlers are thin adapters; future agent tools are equally thin llm.DefineTool[Args] adapters over the same methods — so agent calls inherit permission enforcement for free. Bulk ops designed for agents from the start, e.g. FillRegion(ctx, actor, objectID, region, plantID, spacingOverride) with region a rect/circle in object-local coordinates.
Auth
- OIDC (primary; Authentik).
github.com/coreos/go-oidc/v3+golang.org/x/oauth2(pure Go). Authorization-code flow with PKCE +statecookie; issuer discovery fromPANSY_OIDC_ISSUER; redirect URI derived fromPANSY_BASE_URL. JIT provisioning on first login (the IdP gates access, so the registration flag doesn't apply); the email claim links to an existing local account if one matches, otherwise a user is created and stamped withoidc_issuer/oidc_subject. - Local (fallback). argon2id (
x/crypto/argon2, id variant),subtle.ConstantTimeCompare. Disable entirely withPANSY_LOCAL_AUTH=falsefor pure-Authentik deployments. - Both paths end in the same pansy session cookie.
GET /auth/providerstells the login page which methods to show.
Frontend layout
React 19 + TypeScript + Vite + Tailwind 4 (@tailwindcss/vite), @tanstack/react-router, @tanstack/react-query, zod for API parsing; dev proxy /api → Go server.
- Routes:
/login,/register,/gardens(list),/gardens/:id(editor,?focus=objectId),/plants(catalog). Auth guard on the router root via/auth/me. - State: TanStack Query for all server state (editor keyed on
gardens/:id/full; optimistic mutations with version-conflict rollback). One small Zustand store for ephemeral editor state only: viewport, selection, focused object, active tool, in-flight drag. - Editor components (
web/src/editor/):GardenCanvas(svg root + viewport g),useViewport(use-gesture pan/zoom/pinch),ObjectShape,PlopMarker(semantic-zoom branching),Palette(drag-to-place object kinds),EditorRail(the one side panel),Inspector,HistoryPanel,PlantPicker. - One rail, tabs inside it. The inspector, history, journal, and later the chat panel all want the same strip of screen; rather than each bolting on its own chrome they are tabs in
EditorRail— so the canvas is one width instead of a different width per panel, and adding a panel is adding a tab. Selecting an object switches to the Inspector tab automatically, so the rail is never something you operate before you can edit; on a phone the same tabs render in the bottom sheet the inspector already used. Pure geometry helpers (local↔world transforms, unit formatting) inweb/src/lib/geometry.ts, unit-tested.
Roadmap
- Scaffold — Go module, config, gin server, healthz, sqlite + migration runner; Vite/Tailwind/router shell; embed.FS single-binary build.
- Auth — OIDC (Authentik) + local fallback, sessions, route guard.
- Gardens — CRUD + list page; establishes service-layer and version/409 conventions.
- Field editor — objects, canvas, pan/zoom/pinch, palette, move/resize/rotate, optimistic sync. Riskiest slice — test on a phone at the end of it, not later.
- Plant catalog — ~30 seeded built-ins + custom plants.
- Plops — focus-zoom, place/move/resize, derived counts, semantic zoom. The core vision lands here.
- Sharing — invite by email, roles, viewer read-only mode.
- Polish — imperial toggle, mobile ergonomics, clear-bed, keyboard nudging.
- Agent seam —
ops.gobulk ops +internal/agentDefineTool wrappers. - Garden assistant — majordomo in-process, Ollama Cloud, streaming chat. Each turn runs inside ONE change set (
source='agent'), so a turn that clears a bed and replants it undoes as one action; that is what makes acting without a confirmation prompt defensible. Bounded by a step cap and a timeout — loop safety, not spend control. Themajordomobuild tag is gone: a tag that keeps the agent out of the binary only earns its keep if you'd ship a build without it, and the agent is the point.
Deliberate v1 limits
- Plop
countderived from area ÷ spacing², explicit override allowed. - Shapes: rect + circle only (polygon reserved in schema).
- Seasons = planted/removed dates.
?year=filters/fullto the plops whose[planted_at, removed_at]interval overlapped that calendar year, so garlic planted in October and pulled in July shows in both — and undated plops show in every year, since everything predating the feature has a nullplanted_at. Deliberately noseasonstable: it would duplicate what the dates already say and create a second source of truth about when something was in the ground. Past seasons are read-only; #46's garden copy is the scenario-planning half. - Emoji plant icons (zero assets); SVG icon set later if wanted.
- No background/satellite image tracing (cheap to add later as a garden background field).
- 409-and-refetch conflict handling; no real-time sync.