Skip to content

fix: publish MCP Registry after crates and npm - #1908

Merged
patchloom-release[bot] merged 1 commit into
mainfrom
fix/mcp-registry-publish-after-packages
Jul 22, 2026
Merged

fix: publish MCP Registry after crates and npm#1908
patchloom-release[bot] merged 1 commit into
mainfrom
fix/mcp-registry-publish-after-packages

Conversation

@SebTardif

Copy link
Copy Markdown
Contributor

Summary

0.16.0 Publish MCP Registry failed: the workflow started on release:published when host created the GitHub Release (~17:10), while crates.io and npm only finished ~17:29. The marker wait (10 minutes) timed out before ownership markers existed.

Fix

  • Call publish-mcp-registry.yml from release.yml after publish-npm and custom-publish-crates succeed
  • Drop the race-prone release:published trigger; keep workflow_dispatch for recovery
  • Lengthen CDN lag wait (60 x 30s) and log HTTP status when markers are missing
  • Docs: REPO-SETUP + mcp-setup

0.16.0 recovery

Manual workflow_dispatch with version=0.16.0 is running now (markers already live).

Test plan

  • Local review of workflow YAML
  • Manual dispatch for 0.16.0 green
  • PR CI green
  • Next release uses the post-npm/crates path

The 0.16.0 MCP Registry job timed out waiting for ownership markers
because release:published fires when host creates the GitHub Release,
about 15+ minutes before publish-npm and crates finish.

Call publish-mcp-registry from release.yml after both package publishes
succeed. Drop the race-prone release trigger; keep workflow_dispatch for
recovery. Lengthen the marker wait and log HTTP status for CDN lag.

Signed-off-by: Sebastien Tardif <sebtardif@ncf.ca>
Copilot AI review requested due to automatic review settings July 22, 2026 17:47

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

@patchloom-release
patchloom-release Bot enabled auto-merge (squash) July 22, 2026 17:47
@patchloom-release
patchloom-release Bot merged commit 4ee64e8 into main Jul 22, 2026
46 checks passed
@patchloom-release
patchloom-release Bot deleted the fix/mcp-registry-publish-after-packages branch July 22, 2026 17:52
SebTardif pushed a commit that referenced this pull request Jul 23, 2026
🤖 I have created a release *beep* *boop*
---


##
[0.17.0](patchloom-v0.16.0...patchloom-v0.17.0)
(2026-07-23)


### Features

* library doc_merge selector and embedder text I/O surface
([#1913](#1913))
([0abfb40](0abfb40)),
closes [#1909](#1909)
[#1910](#1910)


### Bug Fixes

* clarify init skip when AGENTS.md is not written
([#1922](#1922))
([4323811](4323811))
* doc.merge catalog blurb mentions multi-doc selector
([#1917](#1917))
([58e3f7a](58e3f7a))
* keep Permission denied and invalid_input on unreadable sole paths
([#1925](#1925))
([cec22bd](cec22bd))
* load errors keep OS detail without double-wrap
([#1926](#1926))
([8dd6987](8dd6987))
* publish MCP Registry after crates and npm
([#1908](#1908))
([4ee64e8](4ee64e8))
* set error_kind on patch check multi-file JSON failures
([#1924](#1924))
([322486c](322486c))
* sole unreadable path error without double-wrap
([#1916](#1916))
([ec40404](ec40404))
* stop double-wrapping load_text_strict IO errors
([#1918](#1918))
([8df2b48](8df2b48))

---
This PR was generated with [Release
Please](https://github.com/googleapis/release-please). See
[documentation](https://github.com/googleapis/release-please#release-please).

---------

Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: patchloom-release[bot] <291219740+patchloom-release[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants