The auto-scroll effect used bottomRef.scrollIntoView(), which walks the
entire scroll-ancestor chain up to the window and scrolls the page to the
log's bottom on every new line. Scroll the <pre> container directly via
scrollTop instead, so only the log pane moves. Drop the now-unused
bottom marker div and ref.
Use the stretched-link pattern: the card is relative and the title link
gets an after:absolute after:inset-0 overlay so clicking anywhere on the
card navigates to the detail page. Interactive controls (action buttons,
subdomain/Docs links, ProgramActions) are raised with relative z-10 so they
stay independently clickable. Cards also get a hover border for affordance.
Reorder the sidebar so Gateway sits right under Overview. Replace the
Scheduled and Program tables with card grids matching Services: add JobCard
and ProgramCard, convert ScheduledSection to a JobCard grid, and turn the
former ProgramTable into a filterable ProgramCard grid (renamed ProgramList;
kept the search box, dropped the sortable table). Remove the now-orphaned
SortHeader component.
Introduce a Layout shell with a collapsible left sidebar (state persisted
to localStorage) that becomes a hamburger drawer under md. Break the single
Dashboard into routed pages — Overview (summary tiles + node bar), Services,
Scheduled, Programs, Gateway, Mesh — nested under the shared Layout via an
Outlet. Centralize useEventStream in Layout (drop the per-page duplicates),
repoint detail-page back links at their section pages, and remove the now-dead
SectionHeader component / SECTION_HEADERS map (each page owns its header).
Both tables wrapped a w-full table in overflow-hidden, so on narrow
viewports the intrinsic column width (badges, action buttons, mono
schedule) got clipped and the section blew past the screen. Switch the
wrapper to overflow-x-auto and give the table a min-w so it scrolls
horizontally within its bordered box instead.
Services often need their own public origin — the URL a browser loads them
at through the gateway — to populate CORS/WebSocket/secure-context allowlists.
Previously this had to be hardcoded in the app's own config, which silently
broke when the subdomain-routing migration changed the address (openclaw's
Control UI rejected its new origin).
${public_url} resolves at deploy to the service's gateway-facing base URL:
https://<name>.<gateway.domain> when exposed under tls: acme, else the
node-local http://localhost:<port>. It tracks gateway.domain, so a domain
change needs no per-app edit. Jobs (not exposed) don't get it.
Matches the TLS mode as the raw lowercased string == "acme", the same way the
Caddyfile generator does (the TLSMode enum is stale).
The proxy->bool refactor (770574f) missed the shared CLI test fixture, which
still used proxy: {caddy: {path_prefix: ...}} — failing ~25 CLI tests with
'Input should be a valid boolean'. Set it to proxy: True.
Also drop the now-dead 'caddy' entry from _STRUCTURAL_KEYS (no nested caddy
key exists anymore) and refresh a stale proxy.caddy docstring reference.
The dashboard's stack dropdown hardcoded its options (python-cli/fastapi/
react-vite), so a supabase program showed "(none)" — the value had no matching
option. Make the backend the single source of truth instead:
- core: available_stacks() returns sorted(HANDLERS) — the stacks castle has
handlers for.
- castle-api: GET /stacks exposes it.
- cli: --stack choices derive from available_stacks() (drops the duplicated list).
- app: ProgramFields fetches /stacks (useStacks) and renders options from it,
always including the current value so it never silently blanks; StackBadge
gets a Supabase label.
A new backend stack now appears everywhere without a frontend edit.
TLSMode (off/internal/letsencrypt) was only referenced by HttpPublic.tls,
and HttpPublic (the expose.http.public field, still carrying a path_prefix
and a tls defaulting to the deleted internal mode) was referenced by no
config, generator, deploy code, or test — a leftover from the pre-proxy:bool
exposure design. Drop the enum, the model, the public field, and both CLI
shim re-exports.
CaddySpec had only `enable` left (path_prefix/host removed; extra_snippets was
unused), so `proxy: { caddy: {} }` was three levels of nesting for a checkbox —
and it invited drift (supabase silently reverted to proxy.caddy.host). Replace it
with a plain `ServiceSpec.proxy: bool`.
- Model: remove ProxySpec/CaddySpec; `proxy: bool = False`. Readers simplified to
`bool(svc.proxy)` (generator, castle-api summaries, CLI info).
- CLI create / dashboard editor + create form: write `proxy: true` / a checkbox.
- Configs migrated to `proxy: true` (and supabase's stale proxy.caddy.host block
removed).
- Docs (registry, dns-and-tls, design, stack guides) + tests/fixtures updated;
also fixed lingering path-model staleness in the stack guides.
Each route address is a subdomain label now, so render it as a link to its full
https://<subdomain>.<domain> URL (via subdomainUrl), opening in a new tab. Falls
back to plain text on a bare host (off mode) where no subdomain URL exists.
The HTTP health poller marked anything >=300 as down. With subdomain routing the
gateway's :<port> health path returns 302 (redirect to the dashboard subdomain),
so castle-gateway showed "down" despite being healthy. Accept <400: a 2xx or 3xx
means the server answered; only 4xx/5xx (or a connection error) is down.
The dashboard is now served at castle.<domain> (was castle-app.<domain>). Since
the subdomain is the program name, this is just a program rename plus the one
special-case constant.
- Generator: _DASHBOARD = "castle" (the :9000 redirect target and off-mode root).
- Program file renamed (runtime): programs/castle-app.yaml -> programs/castle.yaml
(source unchanged: repo:app).
- Dashboard UI: ProgramDetail shows a frontend's subdomain (via subdomainUrl),
dropping the old castle-app-at-root special-case; comment refs updated.
- Docs: registry.md/design.md/react-vite.md updated to the subdomain model and the
`castle` name (react-vite.md also corrected — frontends serve at their subdomain
root with VITE_BASE=/, and builds are manual via `castle program build`, not
deploy).
Follow the backend to the subdomain-only model:
- Service editor + create form: replace the Proxy path / Proxy host text fields
with a single "Expose" checkbox (writes proxy: { caddy: {} } or removes proxy).
- ServiceSummary/DeploymentSummary type: proxy_path/proxy_host → subdomain.
- ServiceCard / ServiceDetail: show the subdomain, linking to
<subdomain>.<domain> via a new subdomainUrl() helper (domain derived from the
dashboard's own host; null on a bare host / off mode).
Collapse the two proxy shapes (path_prefix + host) to a single checkbox: a
service is either port-only, or exposed at <service-name>.<gateway.domain>. Path
routes and their prefix-stripping foot-guns are gone; the subdomain is always the
service name (rename the service to change it).
- Schema: CaddySpec is just `enable` (presence = expose). service_proxy_targets
returns (expose, port, base_url); Deployment carries `subdomain` (was
proxy_path/proxy_host). Registry/deploy/mesh/CLI updated in lockstep.
- Generator: compute_routes emits one host route per exposed service/frontend
(address = name). acme mode = one *.<domain> site with a reverse_proxy matcher
per service + a file_server matcher per frontend; the :<port> site redirects to
the dashboard. off mode = a HTTP control plane on :<port> (dashboard at / +
/api → castle-api); other services are port-only.
- Dashboard/API: castle-app and castle-api are ordinary subdomains. The dashboard
derives the API base at runtime (castle-api.<domain> on a subdomain, else /api)
and calls it cross-origin — castle-api already allows CORS *. Frontends build
with VITE_BASE=/ (served at their subdomain root).
- CLI: `service create` drops --path/--host; --no-proxy makes it port-only.
- Docs/tests reworked for the model; docs use generic placeholders only (no
personal host/domain/IP details).
acme mode fully replaced the internal-CA path (which required installing a private
CA on every device — the exact pain, esp. on Android, that acme avoids). Remove
`internal` entirely and simplify how services declare host routes.
- Generator (caddyfile.py): drop the `tls internal` branch — modes are now
`off` | `acme`. In acme mode the published subdomain is the **first DNS label**
of `proxy.caddy.host` (a bare `claw`, or a legacy `claw.civil.lan`, both →
`claw.<domain>`), so services stay domain-agnostic and the declared value is
authoritative again (no more silent service-name override). Shared
`_host_matcher_block` reused by off-mode and the acme wildcard site.
- castle-api: delete `GET /gateway/ca.crt`, `_gateway_ca_pem`, `_ca_fingerprint`
and the now-unused imports; drop `ca_fingerprint` from `GatewayInfo` (keep
`tls`).
- Dashboard: remove the CA-cert download button + unused imports; drop
`ca_fingerprint` from the `GatewayInfo` type.
- Tests: replace TestCaddyfileTlsInternal with an off-mode class (keeps the
runner-agnostic host-route coverage); acme tests assert first-label derivation
incl. label-wins-over-service-name; drop the castle-api CA-endpoint test.
- Docs: registry.md + dns-and-tls.md — two-mode tables (off|acme), remove the
internal sections/CA-download, document the bare-label host convention; note a
domain-less node stays on `off`.
A conceptual companion to registry.md's field-level gateway reference. Covers the
two questions (resolve / trust) an internal-only LAN service must answer:
- DNS: gateway-as-ingress, path-prefix vs host routes, LAN name resolution,
wildcard records, split-horizon (internal names, no public A records), and the
private-TLD vs public-subdomain zone styles with the worked router/wild-central
topology.
- TLS: the three gateway.tls modes (off / internal / acme) with when-to-use, the
internal-CA distribution pain vs the acme DNS-01 "no client setup" win, the
secure-context (WebCrypto) rationale, the put-a-service-on-HTTPS recipe,
operational prerequisites (443/80 bind, plugin caddy, DNS token, staging), and a
decision table.
Linked from CLAUDE.md's docs list.
internal-CA HTTPS forces every device to trust a private root, which breaks down
on some clients (Android browsers). acme mode instead obtains a real Let's Encrypt
wildcard cert (*.<domain>) via a DNS-01 challenge, so browsers trust it with zero
CA install — while services stay internal-only (no inbound exposure, no public A
records; only a transient _acme-challenge TXT touches the public zone, and LAN DNS
resolves the names to a private IP).
- Config: GatewayConfig/NodeConfig gain domain, acme_email, acme_dns_provider
(plumbed through load/save + registry + deploy, omitted-when-empty for
round-trip stability). tls stays the discriminator: off | internal | acme.
- Caddyfile generator: acme branch emits a global {email, acme_dns <provider>
{env.TOKEN}} block + one *.<domain> wildcard site with host matchers derived
from the service name (<name>.<domain>); path/static stay on :<port>. Shared
_host_matcher_block helper (reused by off-mode). CASTLE_ACME_STAGING=1 toggles
the LE staging CA; acme-without-domain falls back to HTTP with a warning.
- deploy(): _acme_preflight warns (never writes) when domain, the gateway-service
token env, or the CLOUDFLARE_API_TOKEN secret is missing.
- install.sh: opt-in --with-dns-plugin[=provider] builds a DNS-plugin Caddy via
xcaddy (pinned) to /usr/local/bin/caddy, which the gateway picks up on deploy.
- Tests: TestCaddyfileTlsAcme (global block, wildcard site, derived host matcher,
path routes on :port, staging toggle, no-domain fallback). Docs: gateway.tls
modes table + full acme/DNS-01 section (setup, wild-central LAN DNS, staging).
Adds "a stack whose default is a substrate": a shared self-hosted Supabase
backend plus a `--stack supabase` that scaffolds per-app projects (migrations +
edge functions + static UI) which deploy against it. Apps own their code and
stay repo-durable; only their rows/blobs live on the shared substrate.
- compose runner: new RunCompose supervises a multi-container stack as one
systemd unit (ExecStart=`compose up`, generated ExecStop=`compose down`);
secrets via EnvironmentFile. Reusable beyond Supabase. Deployment.stop_cmd
carries the teardown command through the registry.
- supabase stack: CLI --stack choice, STACK_DEFAULTS→frontend with
build.outputs=[public], _scaffold_supabase() (migrations/RLS/functions/
static UI/app manifest), and SupabaseHandler with a forward-only idempotent
migration runner (plan_migrations + psql build) and deno dev-verbs.
- docs: docs/stacks/supabase.md, registered in CLAUDE.md; compose runner
documented in registry.md.
- tests: compose run/stop cmd, systemd ExecStop, compose host-route TLS,
migration planner, supabase verb resolution, create --stack supabase.
When the gateway serves internal-CA HTTPS (gateway.tls=internal), other devices
must trust Caddy's root CA to accept *.lan HTTPS. Surface that from the dashboard
instead of hand-copying root.crt around.
- API: GET /gateway/ca.crt returns the public root cert (PEM) as a download,
sourced from Caddy's admin API (/pki/ca/local — matches the running gateway)
with an on-disk root.crt fallback. 404 unless tls=internal. GatewayInfo gains
`tls` and `ca_fingerprint` (SHA-256, for out-of-band verification). The CA
private key is never exposed — only the public root.
- UI: a "CA cert" button in the Gateway panel, shown only when tls=internal,
linking to the download with the fingerprint in its tooltip.
- Tests for the tls=off defaults and the 404-without-internal-tls path; docs note
the button + endpoint.
The gateway was HTTP-only by design, but a plain-HTTP origin is not a browser
"secure context", so apps that need WebCrypto (device identity, E2E crypto)
can't run when proxied — only localhost or HTTPS qualifies.
Add an opt-in `gateway.tls: internal`. When set, each host route
(proxy.caddy.host) is emitted as its own `<host> { tls internal; reverse_proxy }`
site, served over HTTPS by Caddy's local CA (Caddy binds :443 and redirects :80).
Path-prefix/static routes stay on the HTTP :<port> site — give a service a host
route to put it on HTTPS. Default (unset/off) keeps the existing HTTP-only
Caddyfile (single :<port> block, auto_https off, host matchers inline).
Plumbing mirrors gateway_port: GatewayConfig.tls (castle.yaml load/save) →
NodeConfig.gateway_tls (registry load/save, set in deploy) → the Caddyfile
generator. Tests cover both modes; docs/registry.md documents the option plus
the two operational requirements (binding 443/80 via the unprivileged-port
sysctl, and trusting Caddy's root CA on clients) and the host-route-vs-prefix
guidance that motivates it.
`_require_repo()` now returns the validated repo root (Path) instead of
None, so callers get a non-None path and pyright stops flagging `root /
spec.source` and the `CastleConfig(root=root)` construction. Drops the
redundant `get_castle_root()` re-fetch after each guard and hoists the
`pathlib.Path` import to module scope.
Editing a service's routing (port, path_prefix, host) and then hitting
"apply" or "gateway reload" — rather than a full deploy — left the running
Caddyfile reflecting the old routing. The Caddyfile was generated from the
derived registry.yaml snapshot, which only `deploy()` rebuilds from
castle.yaml; apply/reload regenerated from that stale snapshot.
Make castle.yaml the single source of truth for local routes:
- caddyfile.py: new shared `service_proxy_targets()` derives a service's
(proxy_path, proxy_host, port, base_url) from its spec. `compute_routes`
now builds local routes from `config.services` when castle.yaml is
loadable, falling back to `registry.deployed` only when it isn't. Since
deploy/apply/reload all funnel through `compute_routes`, every regeneration
path now tracks the spec. Mesh/remote and static-frontend routes unchanged.
- deploy.py: `_build_deployed_service` uses the same shared deriver, so the
written registry and the computed routes can't diverge.
- config_editor.py: `/config/apply` now runs a full `deploy()` (rebuilding
units + Caddyfile and reloading the gateway) before restarting services, so
changed unit ExecStarts actually take effect. Scheduled jobs are no longer
restarted on apply. Drops the now-unused direct caddy-reload path.
- Regression tests: a stale registry's old port/path loses to castle.yaml;
registry fallback still works when config is unavailable.
- Introduced `_write_castle_config` function to scatter nested config into separate YAML files for programs, services, and jobs.
- Updated `castle_root` fixture to utilize the new config writing method.
- Added tests for configuration aggregation and scattering in `test_health.py`.
- Removed legacy unit handling from `manifest.py` and refactored related code in `config.py`.
- Updated documentation to reflect new configuration directory structure.
- Removed obsolete unit tests related to unit expansion.
New lines auto-scroll the view to the bottom; scrolling up to read now
pauses that (an amber 'paused' marker shows), and scrolling back to the
bottom resumes — the explicit Pause/Play button and the scroll position stay
in sync via a single at-bottom check. While paused, logs still stream in
below without moving the view. Jump-to-latest is instant rather than smooth
so mid-animation positions never register as a false scroll-up.
A python-runner service ran from a tool venv (uv tool install --editable),
separate from the project venv that uv sync manages. The editable link kept
source live, but dependencies were resolved independently in each venv, so a
new dependency landed in the project venv and never reached the running
service — leaving no clean rollout for it.
Deploy now emits ExecStart=uv run --project <source> --no-dev <program>,
running the service in place from its own project venv. uv run syncs the env
to the lockfile before launching, so a single restart picks up both code and
dependency changes; the ExecStart is deterministic from source and never goes
stale. The tool-venv install is gated to the command runner only; uv tool
install is reserved for tool-behavior programs, where PATH is the point. A
python service with no resolvable source falls back to a PATH lookup.
pnpm 10+ runs a deps-status check before `pnpm <script>` that wants to
purge and reinstall node_modules, aborting in a non-TTY build with
ERR_PNPM_ABORTED_REMOVE_MODULES_DIR_NO_TTY. It is a false positive for a
build verb (which builds, not installs), so ReactViteHandler now prepends
--config.verify-deps-before-run=false to every pnpm call, building against
the existing modules. CI=true is kept as a general non-interactive guard.
The LogViewer header now has a maximize toggle (and a clear button) — maximize
expands it to a full-screen overlay (Esc to restore), so you can read service/
job logs without the 24rem cap. Consolidated the per-page 'Logs' Clear button
into the viewer header (both service and job detail pages).
Generated systemd units hardcoded a thin PATH (~/.local/bin + system dirs)
while dev-verb builds used a separate _EXTRA_PATH_DIRS list, so a service
could be built and run with different tools on PATH. Hoist a single
USER_TOOL_PATH_DIRS in config.py, consumed by both stacks._build_env
(build) and the systemd unit generator (runtime).
Also prefer pnpm's modern $PNPM_HOME/bin shim over the bare $PNPM_HOME dir
(older installs leave a stale version wrapper there) and add
/usr/local/go/bin so Go-based builds/services resolve their toolchain.
Editing a service/job only persisted to castle.yaml; the unit isn't
regenerated and the process isn't restarted (the Restart button just runs
'systemctl restart' on the *existing* unit), so a change like a new port
appeared saved but never took effect.
The config panel now, after saving a service/job, shows an amber 'not live
yet' banner with an Apply button that runs POST /deploy {name} (regenerate
unit + Caddyfile) then, for services, POST /services/{name}/restart — so the
edit actually goes live. Programs are unaffected (no deploy/restart needed).
App builds clean.
The check composite returned ActionResult with no output on success, so the
dashboard's 'Check All' (and the CLI) showed nothing when checks passed — only
a failing sub-verb ever produced output. It now accumulates each sub-verb's
output with a ✓/✗ marker (lint / type-check / test), on both success and
failure, so the output panel always shows what ran.
Also fixes lan-info's type-check: socket.getaddrinfo sockaddr is loosely typed,
so str(infos[0][4][0]) makes the address an unambiguous str.
core 94 green; ruff clean; verified POST /programs/lan-info/check returns output.
Native <select> option popups fell back to the OS default (light) background
while inheriting the dark theme's light text — light-on-light, unreadable.
Pin select/option to the card background + foreground text in index.css.
- Add a 'hint' to the shared Field/TextField; annotate the config forms with
short, accurate descriptions of each field:
- program: source (working copy castle builds from), behavior (tool/daemon/
frontend), stack (optional toolchain template), repo/ref (for clone),
system deps (listed for reference, NOT auto-installed), commands.
- service: port (listened-on, health-checked, proxied; map via ${port}),
health path, proxy path vs proxy host, runs.
- job: schedule (cron → systemd timer), runs.
- ProgramFields: the Commands section is now editable (build/test/lint/
type-check/run), so a program with no stack can declare its own dev verbs
by hand — a declared command overrides the stack default. Verified the
commands/build shape round-trips through the config API.
App builds clean.
handle_path /foo/* matches /foo/ but not the bare /foo, so the no-trailing-
slash form fell through to the root catch-all (castle-app) — which then 404'd
because its router has no such route. Hitting http://gateway:9000/power-graph-app
served the dashboard's 404 instead of Power Graph; same for every proxied
service accessed without the slash.
The generator now emits 'redir /foo /foo/' for each path-prefix route (static,
proxy, remote), so the no-slash form canonicalizes to the slash form that
handle_path matches. Caddy's bare path matcher is exact, so /foo/bar is
unaffected. Host routes and the root '/' are unaffected.
Verified: /power-graph-app → 302 → Power Graph; caddy validate clean; core 94 /
api 52 green.
Sort routes by address in GET /gateway and 'castle gateway status' (so the
dashboard + CLI tables read alphabetically). The Caddyfile keeps its
precedence-sensitive order — only the display is sorted.
The gateway does two things — reverse-proxy services and serve static
frontends — but the dashboard/API route table only ever showed the proxy
routes, so 'serving a frontend' and 'proxying a service' looked like unrelated
features and castle-app/power-graph-app were invisible in the route view.
Now there's one concept: a route maps an address (path '/foo' or host 'foo.lan')
to a target of one kind — static (a built dist served by file_server), proxy
(a local service port), or remote (a service on another node).
- core: caddyfile.py gains compute_routes() — the single source of truth for
the route list; generate_caddyfile_from_registry renders it (output byte-for-
byte identical, verified against the live Caddyfile). Adds a GatewayRoute
dataclass.
- api: GatewayRoute model → {address, kind, target, name, node}; GET /gateway
builds from compute_routes (incl. config for static frontends + mesh for
remote), so the table matches what Caddy actually does.
- app: Gateway panel shows Address · Kind · Target for every route (static
frontends + host routes now appear); program detail shows 'Reachable at
/foo/ · served (static)' for static frontends.
- cli: 'castle gateway status' prints the full route table.
- docs: registry.md/design.md/CLAUDE.md describe routes as one concept with
three target kinds.
core 94 / cli 24 / api 52 green; ruff + app build clean; Caddyfile unchanged.
A service/job's env is now exactly its defaults.env — castle injects no hidden
convention vars. Values support ${port}/${data_dir}/${name} placeholders
(resolved at deploy, alongside ${secret:…}), so a program's own env var names
map to castle's computed values without hardcoding.
Why: the auto-injected <PREFIX>_PORT/<PREFIX>_DATA_DIR were a guess at the
program's env names — right for castle-scaffolded services, dead weight for
adopted ones (lakehouse carried two dead vars; notification-bridge/backup jobs
too). They also weren't visible in the config editor (computed at deploy), which
was the source of the 'four env vars but the UI shows none' mystery.
- core: resolve_env_vars gains a context (${port}/${data_dir}/${name});
deploy builds env from defaults.env only — no <PREFIX>_* injection, no
port_env. Removed the port_env field and the dead _env_prefix helper.
- cli: 'service/job create' gains repeatable --env KEY=VALUE (replaces
--port-env); 'program create' scaffolds <PREFIX>_PORT/_DATA_DIR: ${…} for new
daemons.
- app: removed the 'Port env' field; the Environment editor (defaults.env) is
the single place, with a placeholder hint.
- live migration: central-context/castle-api/power-graph/protonmail mapped their
real vars explicitly; lakehouse → just LAKEHOUSED_DAEMON_PORT: ${port}, data
stays in ~/.lakehoused. Verified all services healthy on their ports, dead
vars gone, zero failed units.
- docs: registry.md/design.md/stack guides + findings updated to the explicit
model.
core 94 / cli 24 / api 52 green; ruff + app build clean.
Sweep of accumulated drift unrelated to the CLI reorg:
- Source paths: ~/.castle/code/<name> → /data/repos/<name> (REPOS_DIR), and
source: code/X → absolute /data/repos/X. registry.md source-resolution table
and 'how programs get in' section rewritten for the /data/repos layout.
- Old field name component: → program: in all YAML examples.
- Old term 'component' → 'program' throughout design.md and the stack guides.
- Frontend serving: corrected the copy-to-artifacts/content model to serve-in-
place from the repo (<source>/<dist>), matching Phase 2b; dropped the dead
_copy_app_static reference and the 'runtime never references the source tree'
claim (frontends are the deliberate exception).
- Stale endpoints/commands: /components → /deployments; castle create/add →
castle program create/add in the create.py/add.py docstrings.
CLAUDE.md was already current. cli 24 green; ruff clean.
Names collide across resource types (a program and a service can share a name),
so the CLI no longer resolves bare names. Operations live under the resource
they act on; platform lifecycle and the cross-resource overview stay top-level.
castle program list|info|create|add|clone|delete|run|install|uninstall|
build|test|lint|format|type-check|check
castle service list|info|create|delete|deploy|enable|disable|start|stop|restart|logs
castle job <same verbs> (create takes --schedule)
castle gateway start|stop|reload|status
castle list | status | deploy [name] # cross-cutting
castle start | stop | restart # whole platform (systemd verbs)
Key changes:
- info/delete/list/run are resource-scoped (no more cross-section name match).
- delete only removes the named resource's entry; program delete still blocks
when a deployment references it.
- per-resource start/stop/restart (systemctl on the unit/timer); top-level
start/stop/restart act on the whole platform.
- expose → 'service create' (more general: --program ref, --runner); gained
'job create' (the CLI never had one).
- dropped up/down — bringing things online is the honest 'deploy && start'.
Docs (CLAUDE.md, registry.md, stack guides) and user-facing messages updated.
core 94 / cli 24 / api 52 green; ruff clean. Verified live: program/service/job
list+info scope correctly, service create+scoped-delete round-trips.
The Stage 2 editor split dropped the old confirm() — Remove program/service/job
deleted immediately on click. Restored confirmation on all three (FormFooter
window.confirm with a per-type message).
Removing a program that still has services/jobs referencing it would orphan
their program: ref. Now blocked:
- API: DELETE /config/programs/{name} → 409 'Programs with active jobs or
services cannot be removed. Delete these first: …' (authoritative).
- App: ProgramFields shows the blocked reason instead of a Remove button when
the program has deployments.
- CLI: castle delete refuses a program with referencing deployments not named
the same (which would survive the delete dangling).
Verified live: DELETE program lakehouse → 409. api 52 / cli 24 green; build clean.
CreateDeploymentForm builds a service or job in castle.yaml (PUT /config/...),
then deploys it (POST /deploy {name} → unit + Caddyfile + reload) and, for a
service, starts it (POST /services/{name}/start), then navigates to the new
detail page.
Reachable two ways:
- Program page Deployments section: 'Create service' / 'Create job' buttons,
prefilled from the program (name, program ref, run target, runner).
- Dashboard: standalone '+ Add service' / '+ Add job' (deployments can run
anything, not just castle programs).
Service form covers port / port_env / health / proxy path / proxy host; job
form covers schedule. Name validated against existing services/jobs.
Verified: the configs the form builds validate (PUT 200) and round-trip clean.
App builds; api 52 / core 94 green from prior stages.
- ProgramActions: install/uninstall (activate) shown only for tools and
no-service frontends. Daemons activate via a service/job, so the program
page no longer offers a misleading Install/Uninstall (which secretly
enabled/disabled the service).
- Program page: new Deployments section listing the services/jobs that run
the program, as links (a program → 0-N services/jobs).
- Service page: 'Runs' row from the run_target summary field (the old code
dug into manifest.run, which was empty pre-Stage-1), a Host row for
proxy_host, and a convenience link to the referenced program.
- lifecycle.activate on a daemon with no service now errors toward
'castle expose' instead of silently installing its binary to PATH.
core 94 green; ruff + app build clean.
