fix: discussion label updates truncated to 3 instead of max labels

[Copilot]

update_discussion was silently truncating labels to 3 due to validateLabels defaulting maxCount to 3 when not explicitly provided — despite tryEnforceArrayLimit above it correctly using MAX_LABELS = 10.

Changes

• update_discussion.cjs: Pass MAX_LABELS explicitly to validateLabels, aligning both limit checks:

  // Before — defaults to maxCount=3
  validateLabels(item.labels, allowedLabels.length > 0 ? allowedLabels : undefined);
  
  // After
  validateLabels(item.labels, allowedLabels.length > 0 ? allowedLabels : undefined, MAX_LABELS);

• update_discussion.test.cjs: Added regression test asserting all 5 requested labels are applied (not truncated to 3); extended defaultLabels fixture to include Label5–Label11. Added a test case asserting that requesting more than 10 labels (> MAX_LABELS) is rejected with an E003 error and no label mutations are called.

[mnkiefer]

@copilot Compare to max count and behavior for issue and PR labels, so as to stay consistent.

[Copilot]

@copilot Compare to max count and behavior for issue and PR labels, so as to stay consistent.

Checked the consistency:

• add_labels.cjs / remove_labels.cjs: use maxCount = config.max || 10 and pass it to validateLabels
• update_issue.cjs: uses tryEnforceArrayLimit(labels, MAX_LABELS, …) (MAX_LABELS = 10), no separate validateLabels call

The fix in cc03161 aligns update_discussion.cjs with both patterns by explicitly passing MAX_LABELS (10) to validateLabels, consistent with the rest of the codebase.

[mnkiefer]

@copilot Also add a test case for more than 10 items.

[Copilot]

@copilot Also add a test case for more than 10 items.

Added in e2900ea — a new test case requests 11 labels (> MAX_LABELS = 10) and asserts the result is { success: false, error: "E003: Cannot add more than 10 labels (received 11)" } with no label mutations called. Also extended defaultLabels with Label7–Label11 to support this test.

[Copilot]

Pull request overview

Fixes update_discussion label updates being unintentionally truncated to 3 labels by aligning validateLabels’ maxCount with the existing MAX_LABELS = 10 enforcement.

Changes:

• Pass MAX_LABELS explicitly into validateLabels within update_discussion.cjs to prevent the implicit maxCount=3 default from truncating label arrays.
• Add regression coverage ensuring 5 labels are applied (no truncation) and that requesting >10 labels fails with E003 and performs no label mutations.
• Extend the test label fixture to include Label5–Label11 so the new cases can resolve label IDs.

Show a summary per file

File	Description
actions/setup/js/update_discussion.cjs	Ensures label validation uses MAX_LABELS (10) instead of validateLabels’ default (3).
actions/setup/js/update_discussion.test.cjs	Adds regression tests for the 10-label limit and prevents silent truncation to 3.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 2/2 changed files
• Comments generated: 0

[mnkiefer]

@copilot Investigate failing CI checks/tests.

[Copilot]

@copilot Investigate failing CI checks/tests.

The CI failure is pre-existing and unrelated to this PR. My changes only touch actions/setup/js/update_discussion.cjs and update_discussion.test.cjs (JavaScript files) — zero Go files changed across all 3 commits.

The failing job is Integration: Workflow Misc Part 2 → TestStepOrderingValidation_SecretRedactionBeforeUploads in pkg/workflow/step_order_validation_integration_test.go. It fails because the compiled workflow still emits a step named "Upload Safe Outputs Items", which satisfies the test's strings.Contains(contentStr, "name: Upload Safe Outputs") check — the test expects that step to have been fully removed (merged into the unified agent artifact). This reproduces identically on the base commit before any of my changes.

[github-actions]

Hey @Copilot 👋 — great catch on the validateLabels / MAX_LABELS mismatch in update_discussion.cjs! This is a clean, well-scoped bug fix and the PR is in excellent shape. Here's a quick summary of how it checked out against the contribution guidelines:

• ✅ On-topic — Directly addresses a bug in a core agentic-workflow action (update_discussion).
• ✅ Follows process — Authored by the Copilot coding agent as expected per CONTRIBUTING.md, with a core team member (mnkiefer) assigned as reviewer.
• ✅ Focused — Touches exactly two files, both scoped to the same bug.
• ✅ No new dependencies — Pure logic fix; no manifest changes.
• ✅ Tests included — Two new regression tests cover the happy path (≤10 labels accepted) and the error path (>10 labels rejected with E003).
• ✅ Clear description — Body explains the root cause, the before/after code, and what the tests assert.

This PR looks ready for maintainer review. Nice work! 🚀

Generated by Contribution Check · ● 2.2M · ◷