Codex runtime mapping

Use the connectors and read-only scopes actually available in this Codex task. Preserve the seven-category coverage map, source investigators, null results, epistemic framework, citation checks and synthesis. Native model/effort follows the standing model-capability-router; delegation does not grant source or write authority. The pinned Cursor skill says readonly mode strips MCP access; current official Cursor Subagents documentation describes readonly as restricting writes and separately says local subagents inherit MCP tools. Treat the source claim as unverified policy, not a platform contract or a reason to grant Codex write access. Before each connector query, verify read-only access in this task; do not request write scope, relay credentials, or send messages. Treat denial as unavailable evidence, not permission to switch tools or infer the answer.

Why

Investigate the motivation and intent behind code. The companion how skill answers what the code does and how it works. why answers what forces led to its shape.

For each delegation, choose the model and effort through the standing model-capability-router and the native Codex inventory. Respect the operator's selection. If a requested model is unavailable, report the exact gap; do not invent a replacement slug.

Operating Posture

Operate as a careful, cautious, and precise investigator. Be honest about what you know vs what you're inferring. Read references/epistemics.md for the full confidence framework and phrasing guide. The synthesizer must follow it.

Step 1. Understand the Target and the Question

The target is usually a chunk of code, a pattern, a feature, or a named design decision. The question is usually a design rationale, a tradeoff, a motivating edge case, an external constraint, dead code, or a broad history sweep.