---
name: "evaluate-issue"
description: "Multi-model evaluation flow for issues, ideas, and architectural decisions"
---
<!-- GENERATED by claude-power-pack - scripts/codex-skill-sync.py; edit .claude/commands/evaluate/issue.md instead -->

## Codex harness adaptations

Generated from a Claude Code command. Where the procedure references these Claude-only surfaces, adapt as follows:

- `AskUserQuestion` tool: ask the user directly in the conversation and wait for their answer.
- MCP tools: use the MCP servers configured in `~/.codex/config.toml`, or fall back to the referenced repo scripts and CLI entry points.
- Helper scripts referenced as `scripts/<name>` are bundled under `scripts/` in this skill directory (byte-identical copies from the claude-power-pack checkout); some expect sibling repo resources, so prefer a full checkout when one is available.
- Canonical guidance linked as `docs/<path>` is bundled under `docs/` in this skill directory (byte-identical copies from the claude-power-pack checkout). The source-relative `../../../docs/...` link in the command body resolves outside an installed skill, so the generated copy points at the bundled path instead. docs/ remains the only writable source.

# Evaluate Issue

Multi-model evaluation flow for issues, ideas, and architectural decisions

## Full procedure

Read `reference.md` in this skill directory for the complete, authoritative procedure before acting on this skill.