opschain actions list

Included from Basic tier

Lists the action catalog. The lane follows the target, the way every catalog verb's does: by default (or with --local) the listing is compiled from the working repository's action files — cheaply, nothing executes — while --server <url> (or the active context) reads one asset's stored catalog from the server.

Locally

• Without --asset, every template's actions are grouped first, then the shared .opschain/actions entries (source repo).
• With --asset (which requires --env), the listing is that asset's effective catalog in resolution order: its template's actions first, then the shared entries, with shared entries a template action of the same name overrides marked shadowed in the STATUS column.
• An action file that fails validation (schema, charset or file-name/header mismatch) still lists, with invalid: <error> in STATUS instead of being silently dropped.
• A repository carrying an actions.rb (only applicable to existing MintPress customers and those using the MintPress SDK or the packs:mintpress feature) holds dynamic actions the static listing cannot know; the listing says so in a trailing note: line, and --discover opts into merging what discovery finds (nothing executes without it).

The table is ACTION SOURCE INPUTS APPROVALS STATUS SUMMARY. SUMMARY is the resolved one-liner: the action's explicit summary: when it declares one, else the description's first sentence (else its first line), truncated with an ellipsis at 120 characters — so the listing stays one line per action however long the description grows. opschain actions show <name> prints the full description; INPUTS and APPROVALS are yes/no for declared input arguments and approval requirements. -o json emits the catalog document instead (entries plus an actions_rb marker and the note).

Against a server

A remote listing reads exactly one asset's stored catalog (/api/projects/…/environments/…/assets/…/actions), so --project, --env and --asset are all required, every entry's source is server, and --discover is refused — discovery is a local-repository lane. An action file the server's catalog recorded as invalid lists with invalid: <error> in STATUS, exactly like the local lane.

Flags

Flag	Meaning
--project <code>	The project code. Required against a remote server; unused locally (the repository is the catalog).
--env <code>	Resolve the catalog for this environment. Required with --asset, and against a server.
--asset <code>	List the asset's effective catalog in resolution order. Requires --env. Required against a server.
--discover	Also run actions.rb plan-mode discovery and merge the discovered entries. Local only.
--filter <column>[!]=<value>	Keep only rows matching the term (repeatable; terms AND). Columns are the printed table headers lowercased (spaces/dashes fold to _); * wildcards match the whole cell case-insensitively, != negates, commas mean any-of. Applied client-side. Grammar: opschain docs cli.

Exit codes

Code	Meaning
0	Catalog listed.
1	The listing failed — repository config unreadable, server/auth error, or a remote lane refusal (see below).
2	Usage error — bad flags.

Named errors

• --asset requires --env — exit 2.
• listing actions against a remote server reads one asset's stored catalog: pass --project, --env and --asset — a server target without the full asset address; exit 1.
• --discover runs the repository's actions.rb discovery lane and only applies locally — drop it for a remote listing — exit 1.

Examples

List the repository's catalog:

opschain actions list

One asset's effective catalog, shadowed shared entries marked:

opschain actions list --env dev --asset web

The same asset's stored catalog on the server the active context targets:

opschain actions list --project acme --env dev --asset web

See also

• opschain docs cli/actions — the command family.
• opschain learn 2.1 (YAML actions), opschain learn 2.4 (action packs), opschain learn 3.3 (how a template version's actions reach this listing).