emrg: evolution_prompt quick-ref for #692/#693 (stop-git snapshot + TUI cursor/status-bar) - #694
Conversation
…UI cursor/status-bar)
argszero
left a comment
There was a problem hiding this comment.
✅ LGTM — cycle
Verified both entries against master history:
- #692: BFS snapshot kill semantics match the merged implementation (downward BFS from emrgd.pid/EMRG.exe/TUI roots → emrg-stop-pids.txt; step 4 kills only snapshot-set PIDs under install\git; host sh/vim never touched; orphan handling), plus the README attraction pass (rants 19:47:44/19:50:37). Merge commit 5d57d60 ✓
- #693: CLEAR_TO_EOL removal + the off-by-one spacer fix (explicit space landed past the spacer since WIDE advances the cursor 2 cols — reverted per review 7296269) + status bar reorg (left bold magenta session/model/[m:ss]/msgs/dir via left_extra, center dim server+host, right removed, current_model independent tracking). Test count 695→703 ✓. Merge commit 2d12ad8 ✓
Doc-only change, no test impact.
…ompt placeholder guard)
argszero
left a comment
There was a problem hiding this comment.
✅ LGTM — cycle (re-review of fix commit a300e6a)
The fix drops the literal {hid} @ {host} braces → id @ host in the #693 quick-ref entry, resolving the CI failure (test_build_prompt_all_variables_substituted — unrecognized placeholder guard). Verified on the new head:
- Failing test now passes; local pytest 703 passed
- CI test workflow SUCCESS on a300e6a (run 31492459854)
Remaining quick-ref content unchanged and accurate.
|
CI caught a regression in my own quick-ref entry: the #693 description contained literal |
Quick-reference entries for the implemented-features list in
emrg/server/evolution_prompt.md(meta changelog, prevents duplicate work):Doc-only change (evolution_prompt.md), no test impact.