Fifth occurrence this session of the same anomaly -- see i2mint/py2mcp#16, thorwhalen/artful#12, thorwhalen/skill#11, i2mint/config2py#23 for the full diagnosis. PR #34 migrates ci.yml to the shared i2mint/wads reusable-workflow stub; every hosted-CI run on the branch completes instantly as action_required with zero jobs.
Confirmed pattern across five repos now (all i2mint or thorwhalen public repos): it happens specifically on a repo's FIRST-ever call to i2mint/wads/.github/workflows/uv-ci.yml@master. Every repo already on the stub before this session (astern, ir, contaix, mixing, falaw, lacing) dispatched CI normally when touched.
Likely cause: a one-time, per-repo "approve this external reusable workflow" gate clearable only via the GitHub web UI (Actions tab). No REST endpoint available to clear it from here.
Given the consistency of this pattern, it may be worth a standing fix in wads-migrate itself (e.g. a note in the ci-to-stub output telling the operator to expect and clear this gate) -- flagging for whoever picks this up, not doing it here since it's a cross-repo wads change outside epythet's own scope.
PR: #34 (left open, not merged).
Fifth occurrence this session of the same anomaly -- see i2mint/py2mcp#16, thorwhalen/artful#12, thorwhalen/skill#11, i2mint/config2py#23 for the full diagnosis. PR #34 migrates ci.yml to the shared i2mint/wads reusable-workflow stub; every hosted-CI run on the branch completes instantly as action_required with zero jobs.
Confirmed pattern across five repos now (all i2mint or thorwhalen public repos): it happens specifically on a repo's FIRST-ever call to i2mint/wads/.github/workflows/uv-ci.yml@master. Every repo already on the stub before this session (astern, ir, contaix, mixing, falaw, lacing) dispatched CI normally when touched.
Likely cause: a one-time, per-repo "approve this external reusable workflow" gate clearable only via the GitHub web UI (Actions tab). No REST endpoint available to clear it from here.
Given the consistency of this pattern, it may be worth a standing fix in wads-migrate itself (e.g. a note in the ci-to-stub output telling the operator to expect and clear this gate) -- flagging for whoever picks this up, not doing it here since it's a cross-repo wads change outside epythet's own scope.
PR: #34 (left open, not merged).