[GH-ISSUE #592] feat: Add multi-app orchestration manifest (apps.json) #167

New issue

Closed

opened 2026-02-26 21:33:52 +03:00 by kerem · 1 comment

kerem commented

2026-02-26 21:33:52 +03:00

Owner

Originally created by @rudrankriyam on GitHub (Feb 17, 2026).
Original GitHub issue: https://github.com/rudrankriyam/App-Store-Connect-CLI/issues/592

Summary

Add a simple way to run the same asc operation across multiple apps (a portfolio), without writing bespoke shell loops.

This should be designed for CI/agents and stay non-interactive.

Why this matters

Many teams manage:

multiple white-label apps
multiple bundle IDs per product
iOS + macOS variants

Running “the same thing” per app today requires custom scripts around ASC_APP_ID / --app.

Current state (verified)

Some commands have their own --all-* flags (domain-specific), but there is no cross-app orchestration.
No .asc/apps.json (or equivalent) manifest exists.
Workflow orchestration is being discussed separately (#581).

Proposed UX

Manifest file

.asc/apps.json (proposal):

{
  "apps": [
    {"name":"App A","id":"123","bundleId":"com.example.a"},
    {"name":"App B","id":"456","bundleId":"com.example.b"}
  ]
}

Command runner

# Run a command for every app in the manifest
asc apps run --file .asc/apps.json -- \
  builds latest --output json

Alternative UX (if you prefer pairing with workflows):

extend asc workflow run (#581) with --apps-file .asc/apps.json

Flags (proposal):

--file (manifest path; default .asc/apps.json)
--concurrency (bounded; default 4)
--continue-on-error (default true; aggregate failures)
--output json|table|markdown (default json)

Output model

Return a deterministic array of per-app results:

app metadata
exitCode
stdout/stderr capture options (off by default to avoid huge output)
timings

Constraints

Must not introduce interactive prompts.
Must be explicit about what is being run.
Should avoid leaking secrets in aggregated outputs.

Test plan (TDD-first)

cmdtests:
- manifest validation errors return exit usage (2)
- runs a harmless subcommand (asc version or --help) per app deterministically
- concurrency and continue-on-error behavior

Acceptance criteria

A manifest-driven multi-app runner exists.
Output is deterministic and CI-friendly.
make test passes.

Originally created by @rudrankriyam on GitHub (Feb 17, 2026). Original GitHub issue: https://github.com/rudrankriyam/App-Store-Connect-CLI/issues/592 ## Summary Add a simple way to run the same `asc` operation across multiple apps (a portfolio), without writing bespoke shell loops. This should be designed for CI/agents and stay non-interactive. ## Why this matters Many teams manage: - multiple white-label apps - multiple bundle IDs per product - iOS + macOS variants Running “the same thing” per app today requires custom scripts around `ASC_APP_ID` / `--app`. ## Current state (verified) - Some commands have their own `--all-*` flags (domain-specific), but there is no cross-app orchestration. - No `.asc/apps.json` (or equivalent) manifest exists. - Workflow orchestration is being discussed separately (#581). ## Proposed UX ### Manifest file `.asc/apps.json` (proposal): ```json { "apps": [ {"name":"App A","id":"123","bundleId":"com.example.a"}, {"name":"App B","id":"456","bundleId":"com.example.b"} ] } ``` ### Command runner ```bash # Run a command for every app in the manifest asc apps run --file .asc/apps.json -- \ builds latest --output json ``` Alternative UX (if you prefer pairing with workflows): - extend `asc workflow run` (#581) with `--apps-file .asc/apps.json` Flags (proposal): - `--file` (manifest path; default `.asc/apps.json`) - `--concurrency` (bounded; default 4) - `--continue-on-error` (default true; aggregate failures) - `--output json|table|markdown` (default json) ## Output model Return a deterministic array of per-app results: - `app` metadata - `exitCode` - `stdout`/`stderr` capture options (off by default to avoid huge output) - timings ## Constraints - Must not introduce interactive prompts. - Must be explicit about what is being run. - Should avoid leaking secrets in aggregated outputs. ## Test plan (TDD-first) - [ ] cmdtests: - [ ] manifest validation errors return exit usage (2) - [ ] runs a harmless subcommand (`asc version` or `--help`) per app deterministically - [ ] concurrency and continue-on-error behavior ## Acceptance criteria - [ ] A manifest-driven multi-app runner exists. - [ ] Output is deterministic and CI-friendly. - [ ] `make test` passes.

kerem

2026-02-26 21:33:52 +03:00

closed this issue
added the
enhancement
label

kerem commented

2026-02-26 21:33:53 +03:00

Author

Owner

@rudrankriyam commented on GitHub (Feb 20, 2026):

Closing this for now. For the current single-app workflow, repo-local workflows are sufficient. We can revisit multi-app orchestration later if there is clear demand from users/teams.

@rudrankriyam commented on GitHub (Feb 20, 2026): Closing this for now. For the current single-app workflow, repo-local workflows are sufficient. We can revisit multi-app orchestration later if there is clear demand from users/teams.

kerem referenced this issue

2026-02-26 21:34:36 +03:00

[PR #170] [MERGED] audit: fix 7 code quality and security issues #332