Skip to content

The dev lockfile is 51 gems behind its own constraints (json 2.7→3.0, zeitwerk 2.6→2.8, async 2.24→2.37) and Dependabot's monthly bundler batch never arrived #64

Open

Description

@minerva-sky

bundle update on main (1d28f84), touching nothing but Gemfile.lock and honouring every cap already in the Gemfile, moves 51 gems. The ones that matter are runtime, and they're what every fresh gem install agentic already resolves to while CI tests the old numbers:

gem lock today after refresh note
json 2.7.1 3.0.2 major. Transitive, unconstrained, so consumers have had 3.x for a while.
zeitwerk 2.6.16 2.8.3 the loader for the whole gem
dry-schema 1.13.4 1.16.0 plus dry-core 1.0→1.2, dry-types 1.7→1.9, dry-logic 1.5→1.6
async 2.24.0 2.37.0 the Gemfile cap is <= 2.37.0; the lock never moved to it. #42 is about the cap itself, this is just honouring it.
thor 1.3.2 1.5.0 the CLI
logger / ostruct 1.6.0 / 0.6.1 1.7.0 / 0.6.3

Dev side: standard 1.32→1.57 (rubocop 1.57→1.91, so Ruby 3.4 syntax finally lints), rspec 3.13.0→3.13.2 (core 3.13.6), webmock 3.23→3.30, simplecov 1.0→1.3, rake 13.2→13.4. The last sweep was #24 on 2026-08-30 and it was deliberately conservative; nothing has moved the lock since except two CVE patches and #33.

Suite on the refreshed lock (Ruby 4.0.2, this machine):

Dependabot never showed up. #43 merged the grouped-monthly config on 2026-09-17. A config merge triggers an immediate run, and the 1st of the month triggers the scheduled one; neither produced a bundler PR, and the only Dependabot PRs this repo has ever had are the two actions/checkout bumps (#9, #19). With seven explicit gems bumpable inside their constraints, "nothing to do" isn't the explanation. The job log lives only in Insights → Dependency graph → Dependabot, which I can't read from here. Worth a glance when you're on the page; I'm not filing that as its own ask.

What I'd do: one PR, bundle update within existing constraints plus the one-line base64 deletion the refresh forces, CHANGELOG untouched (lockfile isn't shipped). CI's 3.2.4 leg is what gates it; locally that leg's resolution is identical except where a gem's floor is 3.3 (none of the runtime ones are).

Class: deps-minor (L1)
Minerva's lean: Refresh now in one PR; CI testing zeitwerk 2.6 and json 2.7 while consumers run 2.8 and 3.0 is the same blind spot #42 named, and this half needs no cap change to close.

Labels: loop:deps, status:analyzed

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    loop:depsDependency / language upgrade loopstatus:analyzedAnalyzed, awaiting decision

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions