[mashkovd]

Summary

• mctl-agents PR feat: introduce service-template catalog #13 landed dead-letter `in-progress` recovery — closed-unmerged PRs now flip to `rejected` instead of stalling forever.
• Image `1.9.0 → 1.10.0`. Backward-compatible MINOR bump.
• Reference incident: `mctl-openclaw/upgrade-to-2026-4-27` (PR feat(agents): upgrade-to-2026-4-27 mctl-openclaw#15, closed 2026-04-30, manually reconciled in PR chore(agents-state): flip mctl-openclaw/upgrade-to-2026-4-27 to rejected #97 before the fix landed).

Test plan

• ArgoCD sync clean
• Next `mctl-agents-shepherd` cron tick spawns a pod on `1.10.0` (no ImagePullBackOff)
• Discovery now picks up `in-progress` proposals with closed PRs (none currently exist post-chore(agents-state): flip mctl-openclaw/upgrade-to-2026-4-27 to rejected #97 cleanup, but synthesized smoketest fixture verifies)

[mashkovd]

@codex review

[mashkovd]

@copilot review

[Copilot]

@copilot review

Here's my review of the changes in this PR (06768c5):

What changed

• image: ghcr.io/mctlhq/mctl-agents:1.9.0 → 1.10.0 (line 178) — single pinned-tag bump
• Top-level workflows.argoproj.io/description annotation updated to reference 1.10.0 (kept in sync ✓)
• The old "NB: ImagePullBackOff is intentional" placeholder comment replaced with a contextual block describing the 1.10.0 feature (dead-letter in-progress recovery, mctl-agents PR feat: introduce service-template catalog #13, reference incident PR chore(agents-state): flip mctl-openclaw/upgrade-to-2026-4-27 to rejected #97)

No issues found

• Scope is minimal: only the image ref and its associated comments change; no logic, env vars, resource requests, volumes, continueOn settings, or synchronization mutex are touched
• imagePullPolicy: IfNotPresent is correctly retained — appropriate for a pinned tag
• Comment quality is an improvement: removes the confusing "intentional ImagePullBackOff" note and documents why 1.10.0 is needed (dead-letter recovery, incident traceback)
• Tag pinning uses a mutable semver tag (:1.10.0) rather than a SHA digest — this is consistent with the existing pattern throughout the file and repo, so it's not a regression introduced here

Minor observation (not a blocker): Using an immutable image digest (@sha256:…) instead of a mutable tag would harden the deploy against accidental tag rebasing, but adopting that pattern would be a repo-wide convention change, not in scope for this bump.

Overall: LGTM — correct, minimal, well-documented change.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Nice work!

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".