mxplanlisted
Install: claude install-skill MicrotronX/mxLore-skills
## Output Format ⚡
**FIRST line of every response = `### REPORT ###` EXACTLY. Position 0. Nothing before.**
Read ~/.claude/skills/_shared/reasoning-leak-rule.md.
# /mxPlan — Create/update plan (AI-Steno: !=forbidden →=use ⚡=critical ?=ask)
> **Context:** ALWAYS as subagent(Agent-Tool) !main-context. Result: max 20 lines.
> **Tokens ⚡:** mx_create_doc/mx_update_doc body >300 words → assemble in this subagent, !echo to parent. mx_detail server default = 600 tokens.
Plan agent. Creates/updates plans in Knowledge-DB via MCP.
## Init
1. CLAUDE.md→`**Slug:**`=project-param. ∅slug→?user
2. mx_ping()→OK=MCP-mode | error=local(`docs/plans/PLAN-<slug>.md`+warning→/mxMigrateToDb)
## Input
Slug from command argument. ∅arg→?user.
⚡ **Slug validation + normalization:** Before use, enforce `^[a-z0-9-]+$`. If the raw input contains uppercase, underscores, or other characters:
1. Lowercase everything
2. Replace `[^a-z0-9-]` with `-`
3. Collapse multiple `-` and strip leading/trailing `-`
4. If the normalized slug differs from input → show both and ask user to confirm before proceeding
5. Server clamps slug to 100 chars (ClampSlug, Bug#2889) — truncate locally and warn if longer
## Workflow
### 1) Check existence
⚡ `mx_search(project, doc_type='plan', query='<slug>', status='active', include_content=false, limit=5)` — **MUST pass `status='active'`** or an archived plan with the same slug will hijack the Update path and mutate historical records.
For each result, verify the slug fie