Repository navigation
Optimize the issue skill family #1347
Description
Part of #1342 — apply the setup-family optimization recipe to the issue family.
8 skills, 47,006 body tokens in total; 1 over the 200-token always-on budget, 6 over the 5,000-token body budget.
| skill | body tokens (budget 5,000) | always-on tokens, est. (budget 200) | lines (cap 500) |
|---|---|---|---|
issue-stale-sweep |
6,422 ⚠ | ~208 ⚠ | 580 ⚠ |
issue-triage |
8,515 ⚠ | ~176 | 797 ⚠ |
issue-reproducer |
6,548 ⚠ | ~197 | 585 ⚠ |
issue-fix-workflow |
6,177 ⚠ | ~183 | 574 ⚠ |
issue-backlog-stats |
6,137 ⚠ | ~193 | 520 ⚠ |
issue-reassess |
5,668 ⚠ | ~197 | 515 ⚠ |
issue-deduplicate |
4,542 | ~193 | 459 |
issue-reassess-stats |
2,997 | ~156 | 319 |
Body tokens are from docs/mode-economics.md; always-on figures are a chars ÷ 4 estimate of description + when_to_use. ⚠ marks a budget exceeded. Line counts include the generated pre-flight block.
Order of work (per the umbrella):
- Trim the frontmatter of every skill marked ⚠ in the always-on column.
- Split any over-cap body, moving sub-action-specific rules into sibling files byte-for-byte.
- Wording pass on the rest, with headings, golden-rule headlines, code blocks and the pre-flight block kept byte-identical.
- Run each touched skill's eval suite before and after, outside the sandbox.
Record any inconsistencies found but not fixed (eval-coupled wording, stale headings) in the PR description.
Activity
liwenjie200543 commentedon Sep 26, 2026
Taking this one, following the #1342 recipe (same as #1390 for
pr-management): trim the always-on frontmatter first, split over-budget bodies into sibling files byte-for-byte, then a wording pass. Headings, golden-rule headlines, code blocks and eval-extracted sections stay byte-identical.ci(labeler): count only explicit references when passing labels to is…
feat(bitbucket): add guarded cloud PR merge (#1471)
feat(tools): add asf-nexus and wire Nexus staging check into verify-rc (