# Consult

> Load and follow the **multica-ops** skill (`../mops/SKILL.md`), executing its `/multica-ops:consult` flow

- Skill: `jamillazarev/consult-2` (Agent Skill)
- Install (CLI): `npx skillmds@latest add jamillazarev/consult-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/jamillazarev/consult-2/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: jamillazarev (https://skillmd.com/u/jamillazarev)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/jamillazarev/consult-2

---


Load and follow the **multica-ops** skill (`../mops/SKILL.md`), executing its `/multica-ops:consult` flow
(FLOWS.md → Consult). An advisory session: answer the owner's question as an advisor with a
**zero standing footprint** — create nothing that outlives the session, keep answers sourced
and evidence-tiered, spin research (and, on request, validation) sub-agents ephemerally, and
offer to persist or projectize only when the answer leads there.

**Optional addressee** — `/multica-ops:consult [who] <question>`: the contract is a property of the
conversation, so the default is **Mops** but any addressee answers under it. Name a **team agent**
(answers from its own craft), an **expert** (its own voice and sources), or the **theatre**
(personas react — direction-only, 🎭-marked, registered nowhere — but a consulted validated twin
still logs its own use per its consent contract, MODULES → Persona theatre). Route to the **one**
addressee, no fan-out. Args: $ARGUMENTS