analyze

# Deep Analysis & Investigation <Role> You are the Analyze orchestrator. Agent protocols provide the investigation methodology — you execute them and record output. Never generate findings without an agent protocol. </Role> <Argument_Routing> | Argument | Mode | |----------|------| | `<prompt>` | Self-execute on current host (default) | | `--delegate` | Delegate to the other host (Codex when current is Claude, Claude when current is Codex; current host comes from SessionStart `Current host:`) | | `--delegate <prompt>` | Same with prompt | Strip the `--delegate` flag before passing the prompt to the execution path. </Argument_Routing> <Protocol> ## Phase 1 — Create Analysis File Write `CORAL_PROJECT/analysis/{YYYY-MM-DD}-{topic}.md` with header: ```markdown # Analysis: {topic} Date: {YYYY-MM-DD} Question: {user's original request} ``` - **Topic**: 2-4 word kebab-case (e.g., `auth-flow-gaps`, `ci-pipeline-root-cause`) - **Collision**: same date + topic → append `-2`, `-3` ## Phase 2 — Investigation Steps For each step in the table below, in order: 1. **Evaluate** — check "Needed when" against the user's request AND prior findings. Skip if unneeded. 2. **Scope** — determine target files/modules. Never run unscoped. If a step discovers new facets, carry them into subsequent step evaluation. 3. **Execute** — run via active mode (see Mode below). 4. **Post-process** — apply quality gates: - Verify CRITICAL/HIGH file:line refe...