القائمة
Use when you need to render an lmd-* skill phase or companion, or when a render call fails and you must decide whether the gateway is broken or the call was misaddressed.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation. Use when starting creative work that will produce a versioned design spec.
Use when implementing any feature or bugfix, before writing implementation code
Use when creating new skills, editing existing skills, or verifying skills work before deployment
Fans out one focused subagent per independent problem domain (disjoint files, no shared state), then conflict-scans and integrates. Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies.
Executes a .lmd.md plan in this session — the main agent runs each task itself, pauses at batch checkpoints for human review, and closes at a whole-branch final gate. No companions. Use when executing an implementation plan inline with human review checkpoints.
Guides branch completion by verifying tests, detecting the workspace environment, presenting exactly 4 (or 3 for detached HEAD) integration options, then executing the choice with provenance-based worktree cleanup. Use when implementation is complete and you must decide how to integrate the work — merge, PR, keep, or discard.
| name | lmd-rendering-skills |
| description | Use when you need to render an lmd-* skill phase or companion, or when a render call fails and you must decide whether the gateway is broken or the call was misaddressed. |
Every lmd-* skill is a delegation stub. Its body lives in the version-pinned pack
@dastholo/lean-md-skills and renders on demand, one phase at a time. Never read a
body or companion file from disk.
In the lean-ctx addon topology — lean-md aggregated behind the lean-ctx gateway,
the common case and the one this skill exists for — ctx_md_render is NOT a direct
tool. lean-md is a separate stdio MCP server, reachable only through the gateway:
ctx_tools(action="call", tool="lean-md::ctx_md_render",
arguments={"skill": "<skill>", "phase": "<phase>"})
Pass exactly one of phase or companion:
arguments={"skill": "<skill>", "companion": "<name>"}
Behind the gateway a direct ctx_md_render(...) call cannot succeed — lean-ctx holds
no lean-md code. The failure reads like a dead gateway; it is a misaddressed call.
(Wired straight into your MCP client instead — lean-md mcp as its own server —
lean-md exposes ctx_md_render as a plain direct tool; then call it bare. Step 1
below tells you which topology you are in.)
ctx_tools(action="list") — does it show lean-md [stdio, enabled]? Then you
are behind the gateway, everything works, and the call was merely misaddressed.
Fix the call, do not fall back. (No lean-ctx gateway at all, but ctx_md_render
sits in your own tool list? Direct topology — the bare call is correct.)Transport closed? Retry once — sporadic, the gateway respawns the server.Read lean_md_bin and lean_md_skills_dir from .lean-ctx/lean-md/vars.toml,
export LEAN_MD_SKILLS_DIR=<lean_md_skills_dir>, then run
<lean_md_bin> render --skill <skill> --phase <phase> --consumer=ai.
Without lean_md_skills_dir the binary reports PACK_MISSING. Same binary,
byte-identical output.