Skip to content
Back to skills

Eplan Router

ASecurity

Routes any EPLAN Electric P8 request to the right reference and documentation index: classifies the task, then reads the matching part of the eplan-scripting skill before acting, and hands EEC Pro work to the separate eec-pro-development skill. Use for EPLAN C# scripting, the EPLAN API, actions, Remote Client automation, macros, parts databases, and to pick the right P8 doc index (2026 semantic, 2027 keyword). Start here when unsure which EPLAN skill applies.

  • 6 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 18, 2026
ai-agentsc#reactdebugginggitapidatabasedocumentation

Works with

  • cli
  • api
  • mcp

Security analysis

A100/100

Scanned November 18, 2026

npx -y skills add Canary-ECAD/eplan-scripting-skill --skill eplan-router --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Eplan Router?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Eplan Router
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/canary-ecad-eplan-router/badge)](https://www.skillsdirectory.com/skills/canary-ecad-eplan-router)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: eplan-router
description: >
  Routes any EPLAN Electric P8 request to the right reference and documentation
  index: classifies the task, then reads the matching part of the eplan-scripting
  skill before acting, and hands EEC Pro work to the separate eec-pro-development
  skill. Use for EPLAN C# scripting, the EPLAN API, actions, Remote Client
  automation, macros, parts databases, and to pick the right P8 doc index (2026
  semantic, 2027 keyword). Start here when unsure which EPLAN skill applies.
---

# eplan-router — EPLAN skill dispatcher

Entry point for EPLAN work. It fingerprints the product, classifies the task, and
names the **minimal** set of skills to read before acting. It dispatches — it does
not re-teach the APIs.

## When to use

- At the **start of any EPLAN request** — writing or debugging a script, API code,
  a Remote Client app, an EEC Pro typical/configuration/generation pipeline — to
  decide which skill(s) and which doc index apply.
- When the user says "EPLAN" without specifying the product.

**When *not* to use:** once the right skill is loaded and the task sits squarely in
it, work from that skill. Re-route only when the task pivots to the other product.

## 1. Product detection

| Product | File signals | Phrase signals | Skill |
|---|---|---|---|
| **EPLAN Electric P8** | `.elk`, `.ema`, `.emp`, `.edz`, `.zw*`, `*.cs` with `[Start]`/`[DeclareAction]`, `Eplan.EplApi.*` | script, action, API, Remote Client, parts database, pages, macros, symbols, ribbon, Cogineer | `skills/eplan-p8/eplan-scripting` |
| **EPLAN EEC Pro** | `.eox` models, `.jmx` Job Server recipes, EEC project/model dirs | mechatronic model, configuration, formula, variant, condition, generator, Job Server, Form-UI | **separate skill** `eec-pro-development` ([eec-pro-skill](https://github.com/Canary-ECAD/eec-pro-skill)) — load it if installed, otherwise suggest installing it |

Ambiguous: an EEC Pro *generation* task that produces P8 output uses **both**:
eec-pro-development for the model, conditions and Job Server side, and
eplan-scripting for the resulting macros, pages and API side. The boundary is
spelled out in the EEC skill. A `Typical *.xlsm` workbook that you insert into P8
yourself, without EEC in the loop, is P8 work: `eplan-scripting` → `eec-typicals.md`.

## 2. Task → skill

- **Write/fix C# for EPLAN P8** (script, add-in, actions, events, ribbon, context
  menus) → `eplan-scripting` → its `references/script-basics.md`,
  `actions-reference.md`, `core-classes.md`.
- **Read/write EPLAN data** (parts DB, project model, pages, properties, symbols)
  → `eplan-scripting` → `references/api-data-access.md`.
- **Check or fix a parts database in bulk** (XML export/import, import modes,
  Data Portal downloads) → `eplan-scripting` → `references/parts-database.md`.
- **Drive EPLAN from outside** (Remote Client, headless, ports, Cogineer) →
  `eplan-scripting` → `references/remoting.md`.
- **Decode/generate a typical workbook into a P8 project** → `eplan-scripting` →
  `references/eec-typicals.md` (the insertion side). The model and conditions side
  is in `eec-pro-development`, a separate skill.
- **Hangs, timeouts, weird behavior in EPLAN automation** → `eplan-scripting` →
  `references/pitfalls.md` first, always.
- **EEC Pro model/typical/formula/config/Job Server/generation work** → the
  separate `eec-pro-development` skill, which also holds the EEC Pro doc-index map.
- **"Which doc index do I search?"** → §3.

## 3. Doc-index selection

Two public P8 indexes with different strengths — pick deliberately:

| Index | Product / version | Search mode | Use it for |
|---|---|---|---|
| `eplan-rag` · `rag2026.covaga.xyz` | P8 docs 2026 | Semantic (bge, ~57k vectors) | P8 questions phrased without exact identifiers |
| `eplan-wiki-2027` · `rag2027.covaga.xyz` | P8 docs 2027 | Keyword/FTS5 + bm25 | Exact names — actions, classes, methods, error codes |

The EEC Pro index (`eecpro-rag`) ships with the separate EEC Pro skill. All remote
indexes are maintained in [eplan-cloudflare-rags](https://github.com/Canary-ECAD/eplan-cloudflare-rags).

Measured head-to-head: FTS5 wins exact-name lookups; semantic wins when the query
shares no vocabulary with the docs. When both P8 indexes are available and you know
the identifier, hit 2027 first.

All expose `POST /search {"query", "topK"}` + `GET /stats`/`/health` over REST, and
the same content over MCP when the `.mcp.json` servers are configured. The local
`eplan` action server (from `eplan-mcp`, not auto-installed) can additionally
*execute* actions/scripts on a running EPLAN and introspect the API live — when it
is connected, prefer live introspection over any doc index.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.