Issues

skills: SKILL.md starts the next round only on a move, so a body-only fix gets none

dexpace/morphic#1615 · opened

on Nov 7, 2026

skills: later-rounds-read-what-moved lists main's files as a stacked change's

dexpace/morphic#1614 · opened

on Nov 7, 2026

skills: recheck-against-final-tree's Probe diffs from a merge-base the merge itself moved

dexpace/morphic#1613 · opened

on Nov 7, 2026

skills: a round after a nit-only round reads far more than the fixes it follows

dexpace/morphic#1611 · opened

on Nov 7, 2026

skills: redundant-guard and two-mechanisms-one-test disagree on two mechanisms no input separates

dexpace/morphic#1610 · opened

on Nov 7, 2026

skills: "Probe only in that worktree" contradicts the rules that plant and trial-merge elsewhere

dexpace/morphic#1609 · opened

on Nov 7, 2026

skills: no angle says a cost a change claims must be measured where it can occur

dexpace/morphic#1608 · opened

on Nov 7, 2026

skills: no rule has a figure in a PR body name its machine, load and budget

dexpace/morphic#1607 · opened

on Nov 7, 2026

skills: simulate-ci-runs says nothing of a job only Actions can trigger

dexpace/morphic#1606 · opened

on Nov 7, 2026

skills: the review setup does not tell two no-op merges from the wrong fetch review-head-in-own-ref warns of

dexpace/morphic#1605 · opened

on Nov 7, 2026

skills: toolchain.md does not say a script started with & has SIGINT ignored

dexpace/morphic#1604 · opened

on Nov 7, 2026

skills: mutation.md does not say how to stop a plant run, and a timeout group outlives a group kill

dexpace/morphic#1603 · opened

on Nov 7, 2026

skills: no rule says how to read a CI job's log, and gh api reads as empty without a flag

dexpace/morphic#1602 · opened

on Nov 7, 2026

skills: the 150-word cap counts a rule's heading and key, and nothing says so

dexpace/morphic#1601 · opened

on Nov 7, 2026

skills: skills_lint reads a backticked $VAR/path as a path in the tree

dexpace/morphic#1600 · opened

on Nov 7, 2026

skills: a clone then switch keeps an end-of-line conversion git status does not show

dexpace/morphic#1599 · opened

on Nov 7, 2026

skills: coverage-gate-on-final-tree gives no command for a body's statement count

dexpace/morphic#1598 · opened

on Nov 7, 2026

skills: nothing says which copy of the review skill a head behind main reads

dexpace/morphic#1597 · opened

on Nov 7, 2026

skills: ci-is-the-gate and pr-claims-match-evidence disagree on a body naming make targets at an unpushed head

dexpace/morphic#1596 · opened

on Nov 7, 2026

skills: body-states-the-head does not say whether a test plan's before-and-after counts as history

dexpace/morphic#1595 · opened

on Nov 7, 2026

skills: no rule reads a universal claim in a body against the input classes its table omits

dexpace/morphic#1594 · opened

on Nov 7, 2026

skills: no rule says when plants across a check's forms and sites may stop

dexpace/morphic#1593 · opened

on Nov 7, 2026

skills: finding-cost names no pprof column for a share of named functions

dexpace/morphic#1592 · opened

on Nov 7, 2026

skills: the push checklist's finding-ID pattern matches Go locals and fixture IDs

dexpace/morphic#1591 · opened

on Nov 7, 2026

skills: the review setup reads main through origin/main, one ref every worktree shares

dexpace/morphic#1590 · opened

on Nov 7, 2026

Navigation Menu

Repository navigation

Issues

skills: SKILL.md starts the next round only on a move, so a body-only fix gets none

skills: later-rounds-read-what-moved lists main's files as a stacked change's

skills: recheck-against-final-tree's Probe diffs from a merge-base the merge itself moved

skills: a round after a nit-only round reads far more than the fixes it follows

skills: redundant-guard and two-mechanisms-one-test disagree on two mechanisms no input separates

skills: "Probe only in that worktree" contradicts the rules that plant and trial-merge elsewhere

skills: no angle says a cost a change claims must be measured where it can occur

skills: no rule has a figure in a PR body name its machine, load and budget

skills: simulate-ci-runs says nothing of a job only Actions can trigger

skills: the review setup does not tell two no-op merges from the wrong fetch review-head-in-own-ref warns of

skills: toolchain.md does not say a script started with & has SIGINT ignored

skills: mutation.md does not say how to stop a plant run, and a timeout group outlives a group kill

skills: no rule says how to read a CI job's log, and gh api reads as empty without a flag

skills: the 150-word cap counts a rule's heading and key, and nothing says so

skills: skills_lint reads a backticked $VAR/path as a path in the tree

skills: a clone then switch keeps an end-of-line conversion git status does not show

skills: coverage-gate-on-final-tree gives no command for a body's statement count

skills: nothing says which copy of the review skill a head behind main reads

skills: ci-is-the-gate and pr-claims-match-evidence disagree on a body naming make targets at an unpushed head

skills: body-states-the-head does not say whether a test plan's before-and-after counts as history

skills: no rule reads a universal claim in a body against the input classes its table omits

skills: no rule says when plants across a check's forms and sites may stop

skills: finding-cost names no pprof column for a share of named functions

skills: the push checklist's finding-ID pattern matches Go locals and fixture IDs

skills: the review setup reads main through origin/main, one ref every worktree shares