opensession: Skill for Claude Code

Interrogate

Spawn one reviewer per configured model to adversarially review code changes. Each model gets the same prompt and rubric. The adversarial signal comes from model diversity, not assigned personas. Models differ in blind spots, priors, and reasoning patterns. Agreement across models is high-confidence signal; lone-model findings are worth reading but lower confidence.

The deliverable is a synthesized verdict. Do NOT auto-apply changes.

Step 1, Determine Scope

Identify what to review from context:

• If the user points at specific files or a diff, use that
• If on a feature branch, run git diff main...HEAD (or the appropriate base branch) for the full changeset
• If the user's message references recent work, gather the relevant files

Package the diff (or file contents) plus any surrounding context files the reviewers need to understand the code.

Step 2, State the Intent

Before spawning reviewers, state the intent explicitly. What is this code trying to accomplish? Derive this from:

• The user's message
• Commit messages
• PR description if one exists
• The code itself

Write one clear paragraph. Reviewers challenge whether the work achieves the intent well, not whether the intent itself is correct. If you're unsure about the intent, ask the user before proceeding.

Step 3, Spawn Reviewers

Discover the policy-gated Open Session session tools. Launch up to four ask-mode reviewers in parallel with spawn_task. Begin each self-contained brief with /pstack and forbid writes. Use the current session or workspace model preset by default. When the preset exposes several configured supporting model families, assign them deliberately to increase independent judgment; otherwise omit model and label the reviewers A through D.

Never inspect host model configuration or invent model ids. If an explicit model is rejected, omit it and continue with normal inheritance. Report the reduced reviewer count when capacity or tool policy prevents four independent runs.