chore: sync workflow templates

[stranske]

Sync Summary

Files Updated

• agents-guard.yml: Agents guard - enforces agents workflow protections (Health 45)
• check_agents_md_freshness.py: Warns when the generated Orchestrator AGENTS.md playbook section cites stale repo paths or commands
• reference_packs.py: Validates and resolves reference pack configuration for shared runner prompt assembly
• orchestrator_skill.py: Validates and resolves exported Orchestrator skill context for remote Codex lanes
• runner_lib/ (1 files): Shared runner prompt assembly, output parsing, and dispatch debounce helpers
• langchain_client.py: LangChain client builder - multi-provider client with slot-based fallback and configuration
• llm_registry.py: LLM model registry helper - shared slot/model selection and blocked-model enforcement

Files Skipped

• pr-00-gate.yml: File exists and sync_mode is create_only
• ci.yml: File exists and sync_mode is create_only
• renovate.json: File exists and sync_mode is create_only
• cross-repo-smoke.yml: File exists and sync_mode is create_only
• llm_slots.json: None

---

Review Checklist

• CI passes with updated workflows
• No repo-specific customizations were overwritten

Source: stranske/Workflows
Source SHA: 821168725d0568cade5bb9a18fc58cab97657975
Template hash: 591316374281
Sync branch: sync/workflows-591316374281
Consumer repo: stranske/Portable-Alpha-Extension-Model
Manifest: .github/sync-manifest.yml

[coderabbitai]

Warning

Review limit reached

@stranske, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 1 minute and 16 seconds. Learn how PR review limits work.

To continue reviewing without waiting, enable usage-based billing in the billing tab.

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan refill rate.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, the refill rate gradually slows as usage increases. The highest same-day bursts are limited more strictly.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 5464ec3f-f517-4c17-aed8-429c0ff8b6bf

📥 Commits

Reviewing files that changed from the base of the PR and between 0a4e22e and 26bd84c.

📒 Files selected for processing (7)

• .github/workflows/agents-guard.yml
• scripts/check_agents_md_freshness.py
• scripts/orchestrator_skill.py
• scripts/reference_packs.py
• scripts/runner_lib/core.py
• tools/langchain_client.py
• tools/llm_registry.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch sync/workflows-591316374281

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 26bd84cd93

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".

[chatgpt-codex-connector]

Preserve default orchestrator summary inclusion

When .reference/ORCHESTRATOR_SKILL.md has already been generated by an earlier step and assemble_prompt is called without materialize_orchestrator_skill or ORCHESTRATOR_SKILL_SUMMARY_PATH, orchestrator_summary_path remains None, so this condition now skips the existing summary. The previous code always appended the default summary when present, so those prompts silently lose the orchestrator instructions unless every caller is updated to pass the new path explicitly.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Fall back after blocking a compare override

If the primary compare override is blocked (for example LANGCHAIN_MODEL=gpt-4o-mini), this branch drops the whole slot after candidate_slots has already been limited to the first two available providers. In that environment compare mode returns only the later provider instead of using the safe slot default for OpenAI; build_chat_client handles the same case by falling back to the configured slot model, so the multi-client verifier loses a comparator whenever a stale blocked override is set.

Useful? React with 👍 / 👎.