Repository navigation
docs(launchpad): AGENTS.md's gh issue create --parent flag does not exist - #171
Merged
Merged
Conversation
…xist launchpad/AGENTS.md §4 rule 2 and §5 gave agents a copy-pasteable recipe for filing a parented issue, ending in --parent 4. There is no --parent flag in gh 2.93.0 — confirmed absent from `gh issue create --help` — so the documented command aborts at argument parsing and no issue is created. Any agent following the spec literally fails on its first filing (#85). The intent — a real GitHub sub-issue link — is correct and achievable; only the mechanism named was wrong. Replaced with the mechanism that actually works: create the issue, then link it in a second step via the REST sub-issues endpoint, which takes the child's database id (not its issue number) as a typed integer. The replacement `link_child` shell function and its usage note (the POST response is the parent, not the child — verify by listing sub-issues afterwards, not by reading the create response) come directly from the issue's own verified reproduction against #81, #83 and #84. Also fixes the secondary defect named in the same issue: filing an ADR via `gh issue create` does not apply .github/ISSUE_TEMPLATE/05-adr.yml's needs-decision label, since issue-form template labels only apply through the web UI. Added the one-line follow-up (`gh issue edit <n> --add-label needs-decision`) immediately after the sub-issue linking guidance, since both are "the CLI doesn't do what the form implies" defects with the same root cause. launchpad/scripts/test-agents-md-parent-fix.sh: confirms no fenced command example in AGENTS.md still passes --parent to gh, and that the link_child replacement is syntactically valid bash. Confirmed it fails against the pre-fix file (finds the old --parent 4 line, and no link_child function to extract). Verified: $ bash launchpad/scripts/test-agents-md-parent-fix.sh 3 passed, 0 failed Closes #85 Signed-off-by: Serina Mcfall <serina.mcfall@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
launchpad/AGENTS.mddocumented agh issue create --parentflag that does not exist in gh 2.93.0 — its central worked example for parenting a sub-issue aborted at argument parsing. Replaced with the mechanism that actually works (create, then link via the REST sub-issues endpoint), taken directly from the filer's own verified reproduction against #81/#83/#84.Related issue
Closes #85
Issue type
Bug
Agent provenance
Objective
Make
launchpad/AGENTS.md's sub-issue-linking example runnable against gh 2.93.0, and fix the secondaryneeds-decisionlabel defect named in the same issue.Impacted components
launchpad/AGENTS.md(§4 rule 2, §5 "Filing an issue")launchpad/scripts/test-agents-md-parent-fix.sh(new)Approach and rejected alternatives
Used the exact replacement mechanism the issue already verified end-to-end (REST sub-issues endpoint, database id not issue number,
gh api -Fnot-f) rather than re-deriving one — it was tested against real issues (#81, #83, #84) and its output pasted in the issue body, so re-verifying an alternative would have been redundant risk for no benefit.No alternative considered for the
needs-decisionfix — it's a one-line documentation addition of an already-diagnosed gap, not a design choice.Verification
Command run:
Raw output:
Confirmed this fails against the pre-fix file:
Not verified
link_childagainst a real issue myself — reusing the issue's own verified output (run against task: document what each milestone is for #81/adr: observability strategy for the cohort Buzz environment #83/adr: testing methodology — what we test, at what level, and what evidence counts #84, read back and confirmed) rather than creating new throwaway issues to re-prove a mechanism GitHub's REST API docs and the issue's own transcript already establish.launchpad/README.mdbeyond confirming (via grep) it has no--parentmentions to begin with — the issue's own "Impacted components" section already checked this and found nothing to fix there.Security implications
None. Documentation only — no code, deployment automation, or CI is affected.
Escalations
None. The issue fully specified both the defect and the verified fix; no ambiguity to raise.