ba8b4d1ace
Discovered during A0-V execution on learning_ai_clock (2026-05-27).
F16: At least 10 of ~50 published @bytelyst/* packages in the Gitea
registry have unrewritten 'workspace:*' refs in their package.json
dependencies. pnpm install inside Docker fails with
ERR_PNPM_WORKSPACE_PKG_NOT_FOUND because there is no workspace context
inside the container.
Confirmed broken (latest version each):
@bytelyst/auth@0.1.5 → errors=workspace:*
@bytelyst/diagnostics-client@0.1.6 → api-client=workspace:*
@bytelyst/events@0.1.5 → queue=workspace:*
@bytelyst/extraction@0.1.5 → api-client=workspace:*
@bytelyst/fastify-auth@0.1.5 → errors=workspace:*
@bytelyst/fastify-core@0.1.5 → errors=workspace:* ← clock dep
@bytelyst/feedback-client@0.1.6 → api-client=workspace:*
@bytelyst/field-encrypt@0.1.6 → errors=workspace:* ← clock dep
@bytelyst/react-auth@0.1.6 → api-client=workspace:*
@bytelyst/sync@0.1.5 → api-client, telemetry-client=workspace:*
Changes:
- § 0: bump count to 16; add F16 row (Critical severity)
- § 0 Implications: F16 blocks every A0-V; updated rationale
- § 3: insert new Phase A-pre (republish + publish-time guard) before A0
- § 3 A0-V: append blocked-status note linking to clock@0be887288
- § 10 Execution order: renumber; insert A-pre as step 3
- § 11 Risk register: add F16 row
Implementation status:
✅ Step 2 (A0 on clock) — committed in learning_ai_clock@0be887288;
Dockerfile + compose changes correct, end-to-end build blocked on F16
⏸ Step 3 (A-pre) — next; touches common-plat publish flow
⏸ Step 4+ (A0-V retry on clock, then onward) — blocked on A-pre
|
||
|---|---|---|
| .. | ||
| docker-build-optimization-roadmap.md | ||
| getting-started.md | ||
| hermes_dashboard_roadmap.md | ||
| hermes-setup-upgrade-roadmap.md | ||
| operations.md | ||
| remove_user_interactive.md | ||
| repo-map.md | ||
| tooling-status.md | ||