Filing-gate class: ① a defect with a named landing site and a measured reach (class a, a public door).
Body last written 2026-10-02T04:47Z: unblocked. #21285 closed completed when PR #21306 merged as dabd1c5be7. Blockers were re-derived with no new one, and the card was dispatched in the same act.
Routing: the maintainer direct-dispatch channel (session_018gA1pE6eJtwHhqx72G8U9X) routes it to domain:cli; triage grades the priority.
Measured (#21306's os-dev, from an empty cwd on the built entry, @oclif/core 5.1.2)
- Short flags.
packages/cli/README.md → ### Global documents -v, --version and -h, --help.
- Plugin group.
### Plugin Management says "There is no os plugin command group in v1". But os plugin build|sign|publish is registered, and the plugin topic shows in os --help.
What to do (shapes, not a ruling)
- Make
### Global true. Either:
- drop
-h / -v from the README (docs follow the implementation); or
- make them work with
additionalHelpFlags: ["-h"] / additionalVersionFlags: ["-v"] in oclif. That is a small product change, so report the four-axis reasoning for whichever is chosen.
- Correct
### Plugin Management to the registered os plugin commands. Read them off os plugin --help, not off memory.
- Re-read the rest of the README's command tables against
os --help while in the file, and give each mismatch a conclusion in the PR body.
Dedup: 450 objectstack issues and PRs listed over REST, titles and bodies grepped for additionalHelpFlags, additionalVersionFlags, command -h not found, os -h, os plugin build and "no os plugin command group". The only hits are #21306's own report and an unrelated lint PR. No card covers this.
Dedupe words: os -h command not found · README Global short flags · additionalHelpFlags · os plugin build sign publish README
Generated by Claude Code
Filing-gate class: ① a defect with a named landing site and a measured reach (class a, a public door).
Body last written 2026-10-02T04:47Z: unblocked. #21285 closed
completedwhen PR #21306 merged asdabd1c5be7. Blockers were re-derived with no new one, and the card was dispatched in the same act.Routing: the maintainer direct-dispatch channel (
session_018gA1pE6eJtwHhqx72G8U9X) routes it todomain:cli; triage grades the priority.Measured (#21306's os-dev, from an empty cwd on the built entry,
@oclif/core5.1.2)packages/cli/README.md→### Globaldocuments-v, --versionand-h, --help.os -hexits 2 withcommand -h not found, andos -vexits 2 withcommand -v not found. Only--helpand--versionwork.@oclif/coreadds no short help or version flag unless theoclifconfig setsadditionalHelpFlags/additionalVersionFlags, andpackages/cli/package.jsonsets neither.### Plugin Managementsays "There is noos plugincommand group in v1". Butos plugin build|sign|publishis registered, and theplugintopic shows inos --help.What to do (shapes, not a ruling)
### Globaltrue. Either:-h/-vfrom the README (docs follow the implementation); oradditionalHelpFlags: ["-h"]/additionalVersionFlags: ["-v"]inoclif. That is a small product change, so report the four-axis reasoning for whichever is chosen.### Plugin Managementto the registeredos plugincommands. Read them offos plugin --help, not off memory.os --helpwhile in the file, and give each mismatch a conclusion in the PR body.Dedup: 450 objectstack issues and PRs listed over REST, titles and bodies grepped for
additionalHelpFlags,additionalVersionFlags,command -h not found,os -h,os plugin buildand "noos plugincommand group". The only hits are #21306's own report and an unrelated lint PR. No card covers this.Dedupe words:
os -h command not found·README Global short flags·additionalHelpFlags·os plugin build sign publish READMEGenerated by Claude Code