feat(tools): #174 OData ALC probe — design, plan, and Phase 1 tooling

[SamErde]

Summary

Phase 1 of the #174 OData adjudication workstream: the runtime ALC-ownership probe tooling, plus the design spec and implementation plan it was built from. The actual #174 resolution is Phase 2 (evidence-gated, separate).

Design + plan (docs)

• docs/superpowers/specs/2026-06-01-issue174-odata-alc-probe-design.md — the design (probe extension, runbook, workaround-first/preload-last adjudication framework).
• docs/superpowers/plans/2026-06-01-issue174-odata-alc-probe.md — the task-by-task implementation plan.

Tooling (Phase 1)

• Component A — new tools/Get-DLLPickleLoadedTrackedAssembly.ps1: in-session dump of loaded assemblies whose name is in trackedAssemblies, with version + ALC (optional -NameLike post-filter).
• Component B — tools/Get-DLLPickleRuntimeAssemblySnapshot.ps1 refactored to source its filter from trackedAssemblies via the shared helper + a new -PolicyPath param (so it now captures Microsoft.OData.*/Microsoft.Spatial), replacing the hardcoded regex.
• tests/Unit/RuntimeAssemblyProbe.Tests.ps1 — 5 unit tests for A and B.
• docs/Architecture.md §4 — component map lists the new tool.

Validation

• Invoke-Build Analyze,Test — 4 tasks, 0 errors (1 warning = the existing InvalidLibrary.dll negative-test fixture); analyzer clean on src/tests/tools.
• Local smoke: after Import-DPLibrary, the helper reports the 9 preloaded MSAL/IdentityModel assemblies (Default ALC) — end-to-end confirmation.

No bundle paths touched (tools//tests//docs/), so no PSGallery release on merge.

🤖 Generated with Claude Code

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[github-actions]

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

[codacy-production]

Up to standards ✅

🟢 Issues 1 minor

Results:
1 new issue

Category	Results
CodeStyle	1 minor

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[Copilot]

Pull request overview

Adds a design spec for the next workstream to gather runtime AssemblyLoadContext (ALC) evidence for the Microsoft.OData.* stack (focused on Az.Storage + ExchangeOnlineManagement) and to define an evidence-gated adjudication framework for issue #174.

Changes:

• Documents a Phase 1 plan to extend the existing runtime probe to use build/dependency-policy.json’s trackedAssemblies (capturing OData assemblies).
• Specifies an in-session “loaded tracked assemblies” dump helper plus a maintainer-run live-probe runbook.
• Defines a workaround-first / preload-last decision framework for resolving #174 in a follow-up phase.

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 1 comment.

[SamErde]

@codex review

[Copilot]

Pull request overview

Copilot reviewed 6 out of 6 changed files in this pull request and generated no new comments.

Comments suppressed due to low confidence (1)

tools/Get-DLLPickleRuntimeAssemblySnapshot.ps1:71

• pwsh is invoked as a child process, but the script doesn’t check $LASTEXITCODE and will return an empty array if the child terminates before producing JSON (e.g., missing helper/policy path). That can silently mask failures and look like “no tracked assemblies loaded”. Consider throwing on non-zero exit, and wrapping JSON parsing with a clearer error so probe failures are visible to the caller.

    $Raw = & pwsh @ChildArguments
    $Json = ($Raw | Out-String).Trim()
    if ([string]::IsNullOrWhiteSpace($Json)) { return @() }
    @($Json | ConvertFrom-Json)