feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
import { readFileSync } from "node:fs" ;
import { parse } from "yaml" ;
import { z } from "zod" ;
2026-07-14 22:20:27 +00:00
// Coolify validates every checkout-relative path on create and 422s anything
// that is not absolute. Transcribed from coolify v4.1.2:
// `bootstrap/helpers/api.php::sharedDataApplications()` binds
// `base_directory`/`publish_directory` to `ValidationPatterns::directoryPathRules()`
// and `docker_compose_location` to `ValidationPatterns::filePathRules()`, and
// `app/Support/ValidationPatterns.php` defines the two patterns below. The only
// difference between them: a file path needs at least one character after the
// slash, a directory path may be the bare `/` (the checkout root).
const COOLIFY_FILE_PATH = /^\/[a-zA-Z0-9._/~@+-]+$/ ;
const COOLIFY_DIRECTORY_PATH = /^\/[a-zA-Z0-9._/~@+-]*$/ ;
// These are refinements, not normalizations, and must stay that way: cast does
// not quietly rewrite what the manifest says. A value that would 422 gets fixed
// in the file, in a commit, once — not repaired in memory on every run. And the
// check belongs here, at parse time, because it is a property of the manifest
// and of nothing else: by the time a create returns its bare 422, `apply` has
// already made the project and the environment, and the run is half-applied.
const composeFilePath = z
. string ( )
. regex (
COOLIFY_FILE_PATH ,
"compose_file must be an absolute path inside the repo checkout (Coolify 4.1.2 rejects the create otherwise) — write /docker-compose.yaml, not docker-compose.yaml" ,
) ;
const repoDirectoryPath = ( field : string ) = >
z
. string ( )
. regex (
COOLIFY_DIRECTORY_PATH ,
` ${ field } must be an absolute path inside the repo checkout (Coolify 4.1.2 rejects the create otherwise) — write /apps/core, not apps/core; the checkout root is / ` ,
) ;
feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
const AppSpecSchema = z
. object ( {
source : z.object ( { repo : z.string ( ) , branch : z.string ( ) } ) . strict ( ) ,
build : z
. object ( {
pack : z.enum ( [ "nixpacks" , "static" , "dockerfile" , "dockercompose" ] ) ,
2026-07-14 22:20:27 +00:00
base_directory : repoDirectoryPath ( "base_directory" ) ,
publish_directory : repoDirectoryPath ( "publish_directory" ) . optional ( ) ,
compose_file : composeFilePath.optional ( ) ,
fix(apply): express static-site build settings so a monorepo app is served, not run (#63)
apply created applications but dropped install_command, build_command, and
is_static — settings the manifest had no field for — so a static site in an
npm-workspace monorepo (landing) was built and RUN from the repo-root
package.json, booting the core API server, which crash-looped on a missing
DATABASE_URL.
The build block gains install_command / build_command / start_command
(free-form strings) and static (-> Coolify is_static). apply writes and diffs
them; draft emits them (they left its NO_HOME list, and is_static was never in
it — the silent loss that caused the crash), and only emits static alongside a
publish_directory so a draft always loads.
Managing is_static is opt-in: declaring `static:` is required to serve a static
app, and NOT emitting is_static by default avoids the first apply PATCHing
static serving OFF on an un-migrated app (or fighting a pack:static coupling
forever). static:true with no publish_directory, and any of the four on a
dockercompose app, are parse-time refusals.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-14 23:46:14 +00:00
// The three build/run commands and the static flag Coolify accepts on
// both the create (POST /applications/private-github-app) and the
// update (PATCH /applications/{uuid}) routes. They are free-form
// strings passed through verbatim — cast does not parse or validate the
// shell in them, only whether they belong on this pack (superRefine
// below). `static` maps to Coolify's `is_static`: it makes Coolify
// SERVE `publish_directory` and run NO start command, which is exactly
// the fix for a static site in a workspace monorepo that otherwise gets
// built and RUN from the repo-root package.json (#63).
install_command : z.string ( ) . optional ( ) ,
build_command : z.string ( ) . optional ( ) ,
start_command : z.string ( ) . optional ( ) ,
static : z . boolean ( ) . optional ( ) ,
feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
} )
. strict ( ) ,
port : z.number ( ) . int ( ) . optional ( ) ,
healthcheck : z.string ( ) . optional ( ) ,
domains : z.array ( z . string ( ) ) . optional ( ) ,
service_domains : z.record ( z . array ( z . string ( ) ) ) . optional ( ) ,
env_template : z.string ( ) . optional ( ) ,
} )
. strict ( )
. superRefine ( ( app , ctx ) = > {
if ( app . build . pack === "dockercompose" ) {
if ( ! app . build . compose_file )
ctx . addIssue ( {
code : "custom" ,
message : "dockercompose apps require build.compose_file" ,
} ) ;
if ( ! app . service_domains )
ctx . addIssue ( {
code : "custom" ,
message : "dockercompose apps require service_domains" ,
} ) ;
for ( const k of [ "port" , "healthcheck" , "domains" ] as const )
if ( app [ k ] !== undefined )
ctx . addIssue ( {
code : "custom" ,
message : ` ${ k } not allowed on a dockercompose app (lives in the compose file) ` ,
} ) ;
if ( app . build . publish_directory )
ctx . addIssue ( {
code : "custom" ,
message : "publish_directory not allowed on a dockercompose app" ,
} ) ;
fix(apply): express static-site build settings so a monorepo app is served, not run (#63)
apply created applications but dropped install_command, build_command, and
is_static — settings the manifest had no field for — so a static site in an
npm-workspace monorepo (landing) was built and RUN from the repo-root
package.json, booting the core API server, which crash-looped on a missing
DATABASE_URL.
The build block gains install_command / build_command / start_command
(free-form strings) and static (-> Coolify is_static). apply writes and diffs
them; draft emits them (they left its NO_HOME list, and is_static was never in
it — the silent loss that caused the crash), and only emits static alongside a
publish_directory so a draft always loads.
Managing is_static is opt-in: declaring `static:` is required to serve a static
app, and NOT emitting is_static by default avoids the first apply PATCHing
static serving OFF on an un-migrated app (or fighting a pack:static coupling
forever). static:true with no publish_directory, and any of the four on a
dockercompose app, are parse-time refusals.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-14 23:46:14 +00:00
// A compose app builds and runs from its compose file — Coolify never
// consults these on it. Reject them at parse time rather than post them
// and have them silently ignored (the same reasoning as publish_directory
// and port above).
for ( const k of [
"install_command" ,
"build_command" ,
"start_command" ,
] as const )
if ( app . build [ k ] !== undefined )
ctx . addIssue ( {
code : "custom" ,
message : ` build. ${ k } not allowed on a dockercompose app (it builds from its compose file) ` ,
} ) ;
if ( app . build . static !== undefined )
ctx . addIssue ( {
code : "custom" ,
message :
"build.static not allowed on a dockercompose app (a compose file decides what is served)" ,
} ) ;
feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
} else {
if ( ! app . domains )
ctx . addIssue ( {
code : "custom" ,
message : "domains required (non-compose app)" ,
} ) ;
if ( app . service_domains || app . build . compose_file )
ctx . addIssue ( {
code : "custom" ,
message :
"service_domains/compose_file only allowed with pack dockercompose" ,
} ) ;
fix(apply): express static-site build settings so a monorepo app is served, not run (#63)
apply created applications but dropped install_command, build_command, and
is_static — settings the manifest had no field for — so a static site in an
npm-workspace monorepo (landing) was built and RUN from the repo-root
package.json, booting the core API server, which crash-looped on a missing
DATABASE_URL.
The build block gains install_command / build_command / start_command
(free-form strings) and static (-> Coolify is_static). apply writes and diffs
them; draft emits them (they left its NO_HOME list, and is_static was never in
it — the silent loss that caused the crash), and only emits static alongside a
publish_directory so a draft always loads.
Managing is_static is opt-in: declaring `static:` is required to serve a static
app, and NOT emitting is_static by default avoids the first apply PATCHing
static serving OFF on an un-migrated app (or fighting a pack:static coupling
forever). static:true with no publish_directory, and any of the four on a
dockercompose app, are parse-time refusals.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-14 23:46:14 +00:00
// `static: true` tells Coolify to serve publish_directory and run no
// start command — so a static app with nothing to serve is almost
// certainly a mistake, and one that would deploy green while serving an
// empty site. Catch it in the file, once, not on a live box.
if ( app . build . static === true && ! app . build . publish_directory )
ctx . addIssue ( {
code : "custom" ,
message :
"build.static: true serves publish_directory and runs no start command — but no publish_directory is set, so there is nothing to serve" ,
} ) ;
feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
}
} ) ;
const DatabaseSpecSchema = z
. object ( {
type : z . enum ( [ "postgresql" , "redis" ] ) ,
version : z.string ( ) . optional ( ) ,
backup : z
. object ( { frequency : z.string ( ) , retention : z.number ( ) . int ( ) } )
. strict ( )
. optional ( ) ,
} )
. strict ( ) ;
const ServiceSpecSchema = z
. object ( {
type : z . string ( ) ,
domains : z.array ( z . string ( ) ) . optional ( ) ,
env_template : z.string ( ) . optional ( ) ,
} )
. strict ( ) ;
const EnvironmentSpecSchema = z
. object ( {
applications : z.record ( AppSpecSchema ) ,
databases : z.record ( DatabaseSpecSchema ) . optional ( ) ,
services : z.record ( ServiceSpecSchema ) . optional ( ) ,
feat: cast capture — adopt a hand-built Coolify into the age secret store (#15)
cast was scoped to the steady state: manifest → Coolify, forever. It had no
adoption path — no way to bootstrap the age store from an instance built by
hand, before any manifest existed. The operator did it by hand: curl the envs,
assemble 17 name=value pairs into /dev/shm/prod.env, age -r, shred. Every input
to that pipeline is something cast already has, so a human was shuffling cast's
own inputs through a terminal, with the leak (scrollback, history, a tmp file
that never got shredded) and the silent miss both live.
cast capture <org>/<repo> --env <env> [--generated N] [--override N] [--force]
The required set comes from the MANIFEST, not the box: the ${...} refs in that
environment's env templates, read by the same parser apply uses to demand them.
resolveTemplate and templateRefs now share one grammar — a drift between them
would mean capture collects a different set than apply later requires, which is
exactly the "a name silently missed" failure this verb exists to remove.
The mapping is deliberately NOT mechanical. A DATABASE_URL read off the source
points at the SOURCE box's Postgres: confidently wrong, entirely plausible, and
the target's real URL does not exist until Coolify creates the resource. So the
manifest declares `generated_secrets:` and those names are written as the
literal `pending-coolify-generated`. staging's ADMIN_EMAIL must be the operator,
not the source's — staging and prod share a Mailgun domain, so a staging box
carrying the real address can mail real users; that is --override.
A "capture everything" verb would be wrong in ~4 of 17 entries, silently —
worse than being wrong in all of them. So every name is forced into a
disposition, and two of the four stop the run: a name required by a template but
absent from the source REFUSES (an empty substitutes to nothing and the app
boots misconfigured), as does one name carrying different values on two
resources.
generated_secrets is a manifest property rather than a flag the operator must
remember, because the manifest is what knows DATABASE_URL comes from a database
it declares. An entry no template refers to is a hard error: a guard standing
over nothing reads like a guard, and the likeliest cause is a typo whose real
name is then captured from the source instead of placeheld.
Secret hygiene, all covered by tests asserting on real values:
- the plan prints names and provenance, NEVER values
- an --override's value comes from $CAST_CAPTURE_<NAME>, never argv (`ps`)
- plaintext is piped to age on stdin — never a temp file, stdout, or history
- an existing store is not overwritten without --force: it may hold the only
copy of values the source no longer has (apply's never-delete, applied here)
capture inherits diff's absent-target refusal (D-237) — against a project that
isn't there it would report every secret as missing, an alarming report about
the wrong box — plus the team assert and the --path/--env prod ban. The last
gate is a typed confirmation of the environment's name; there is no --yes.
The end-to-end test decrypts the store cast wrote and asserts on its contents,
so "exactly the names the manifest requires, no more and no fewer" is checked
against real ciphertext rather than against cast's own console output.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-07-13 16:51:43 +00:00
// Secret names whose values the PROVIDER generates — a Coolify-created
// Postgres/Redis URL, a service's own generated credentials. `capture`
// writes these as the literal `pending-coolify-generated` and never copies
// the source box's live value: that value points at the SOURCE box's
// database, so carrying it over would be confidently wrong in a way that
// looks entirely plausible, and the target's real URL does not exist until
// Coolify creates the resource.
//
// It is a manifest property rather than a flag the operator has to
// remember, because the manifest is what knows DATABASE_URL comes from a
// database it declares. Optional: a manifest that names none simply has no
// generated secrets, and `capture` will say so in its plan.
generated_secrets : z.array ( z . string ( ) ) . optional ( ) ,
feat: cast — the Coolify executor, extracted from the infra state repo
Public tool, private state. cast holds no hostnames, no bindings, no
secrets: it joins a product repo's .infra/ manifest with a state directory
you point it at, and makes Coolify match.
Extracted from heavy-duty/infra, which was half tool and half state — the
inconsistency that made it impossible to say whether "infra" named a CLI
or a runbook. rig builds the boxes; cast fills them; infra is what they
are filled with.
Two changes were required to make it genuinely stateless and publishable:
- The implicit cwd contract (environments.yaml / secrets/ / .coolify.env
resolved against the working directory, silently reading the wrong file
from the wrong place) is now an explicit --state <dir> / $CAST_STATE.
- BANNED_IN_PROD — a hardcoded list of one product's ALLOW_* flags, the
only product knowledge in the executor — becomes the generic, operator-
owned environments.<env>.forbidden_var_patterns. The guard now lives in
private state, so a product-side change cannot lower its own guard, and
it is a pattern rather than a list, so it catches unforeseen siblings.
Age identities resolve as $CAST_AGE_KEY_FILE_<ENV> then
~/.config/cast/age-<env>.key — which is the entire attended-vs-unattended
apply mechanism, with no environment names known to the tool.
Instance identity (org names, the GitHub App name, founder domains) is out
of the fixtures and out of register-github-app.sh, which took APP_NAME and
ORG as arguments rather than baking them in.
69 tests green; bin/cast + curl installer mirror rig's shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 12:25:44 +00:00
} )
. strict ( ) ;
const ManifestSchema = z
. object ( {
project : z.string ( ) ,
environments : z.record ( EnvironmentSpecSchema ) ,
} )
. strict ( ) ;
export type AppSpec = z . infer < typeof AppSpecSchema > ;
export type DatabaseSpec = z . infer < typeof DatabaseSpecSchema > ;
export type ServiceSpec = z . infer < typeof ServiceSpecSchema > ;
export type EnvironmentSpec = z . infer < typeof EnvironmentSpecSchema > ;
export type Manifest = z . infer < typeof ManifestSchema > ;
export function loadManifest (
path : string ,
opts : { overrideText? : string } = { } ,
) : Manifest {
const text = opts . overrideText ? ? readFileSync ( path , "utf8" ) ;
const result = ManifestSchema . safeParse ( parse ( text ) ) ;
if ( ! result . success ) {
throw new Error (
` invalid manifest ${ path } : ${ result . error . issues . map ( ( i ) = > ` ${ i . path . join ( "." ) } : ${ i . message } ` ) . join ( "; " ) } ` ,
) ;
}
return result . data ;
}