issues: file #51 #52 (external hu13 filings)
Message
issues: file #51 #52 (external hu13 filings)
Diff
.forge/issues/6c6e579d-1ad5-41c6-ae80-9649cf626000/issue.md+15 −0
--- /dev/null+++ b/.forge/issues/6c6e579d-1ad5-41c6-ae80-9649cf626000/issue.md@@ -0,0 +1,15 @@1+---2+id: 6c6e579d-1ad5-41c6-ae80-9649cf6260003+title: '[external:hu13] issue edit should accept --body-file (parity with create)'4+state: open5+labels: []6+assignees: []7+milestone: null8+created: 2026-09-10T09:29:45.539885439Z9+updated: null10+---11+Filing from hu13 (external — do not implement without authorisation).12+13+f0rge issue create takes --body-file; issue edit only takes inline --body. Editing a long issue body (e.g. updating a plan/rubric) from scripts requires shell-quoting the whole body (backticks execute in double quotes) — I had to shell out through python to pass it safely as argv.14+15+Ask: add --body-file F to issue edit (and pr edit if it exists). No behaviour change otherwise.
.forge/issues/76d39ea4-a171-4124-9214-47c5d1f0fcac/issue.md+15 −0
--- /dev/null+++ b/.forge/issues/76d39ea4-a171-4124-9214-47c5d1f0fcac/issue.md@@ -0,0 +1,15 @@1+---2+id: 76d39ea4-a171-4124-9214-47c5d1f0fcac3+title: '[external:hu13] merge-gate failure should point at the failing log'4+state: open5+labels: []6+assignees: []7+milestone: null8+created: 2026-09-10T09:29:52.198643572Z9+updated: null10+---11+Filing from hu13 (external — do not implement without authorisation).12+13+When f0rge merge gates a patch, failure prints only e.g. error: ci fail: 01_clippy.sh with no log excerpt or pointer. I had to recreate the scratch merge by hand and run the step to find the failing lint (a fresh-build-only useless_format that incremental clippy had masked — the gate was RIGHT, which is exactly why the log matters).14+15+Ask: on gate failure, print the failing step's last ~20 log lines, or a pointer to where the scratch run log lives (ci log <run> <step> shape if applicable). Nothing else changes.