config-export

What this skill does

# config-export

> Exports CLAUDE.md and ai-context/ to tool-native instruction files for GitHub Copilot, Google Gemini, and Cursor.

**Triggers**: `/config-export`, export config, copilot instructions, gemini config, cursor rules

---

## Purpose

`config-export` reads the project's Claude configuration (`CLAUDE.md` and any present `ai-context/` files) into an in-context bundle, lets the user select one or more target AI assistants, generates tool-specific instruction files via LLM transformation, shows a dry-run preview, and writes the confirmed files to their canonical locations. It never modifies source files and never writes without explicit user confirmation.

---

## Process

### Step 1 — Source collection

**Guard — no CLAUDE.md:**
If no `CLAUDE.md` exists in the current working directory, present the user with two paths:

```
No CLAUDE.md found in the current directory.

Choose an option:
  1. project-setup first  — run /project-setup to scaffold CLAUDE.md + ai-context/ in this project,
                            then re-run /config-export (recommended for new projects with SDD workflow)
  2. SDD bootstrap         — generate a minimal .github/copilot-instructions.md from the global SDD
                            template only (no project-specific stack or conventions)

Enter 1 or 2 (or press Enter to cancel):
```

- If the user selects **1**: emit `Run /project-setup, then re-run /config-export` and stop. No files are written.
- If the user selects **2**: proceed using the global SDD methodology as the sole source bundle (skip ai-context/ collection; note "No project CLAUDE.md — bootstrap mode" in transformation context).
- If the user cancels: emit `Export cancelled — no files written` and stop.

**Guard — no ai-context/:**
If no `ai-context/` directory is present, emit:

```
WARNING: ai-context/ not found — export quality will be lower; only CLAUDE.md will be used as source
```

Continue; `CLAUDE.md` alone is sufficient.

**Read and bundle source files in priority order:**

| #   | File                         | Required? | Behavior when absent                                                 |
| --- | ---------------------------- | --------- | -------------------------------------------------------------------- |
| 1   | `CLAUDE.md`                  | Yes       | Halt (see guard above)                                               |
| 2   | `ai-context/stack.md`        | No        | Skip; note "stack.md not available" in transformation context        |
| 3   | `ai-context/architecture.md` | No        | Skip; note "architecture.md not available" in transformation context |
| 4   | `ai-context/conventions.md`  | No        | Skip; note "conventions.md not available" in transformation context  |
| 5   | `ai-context/known-issues.md` | No        | Skip silently if absent                                              |

All present files are read into the in-context bundle before any transformation begins.

---

### Step 2 — Target selection

**If the user provided a target as a CLI argument** (e.g., `/config-export copilot`), skip the interactive menu and proceed directly to Step 3 with that target.

**Otherwise, present the target menu:**

```
Available export targets:

  1. copilot  →  .github/copilot-instructions.md
  2. gemini   →  GEMINI.md
  3. cursor   →  .cursor/rules/conventions.mdc
               →  .cursor/rules/stack.mdc
               →  .cursor/rules/architecture.mdc

Enter targets (comma-separated numbers or names, or "all"):
```

**Claude target rejection:**
If the user requests target `claude` or `CLAUDE.md`, respond with:

```
The Claude target is not supported in V1 — use /project-update to refresh CLAUDE.md
```

Do not write any files; re-present the menu or halt based on user choice.

---

### Step 3 — Dry-run generation and confirmation

For each selected target, apply the corresponding transformation prompt (see sub-sections below). Then:

1. Display the full generated content for each file and its destination path.
2. If a target output file already exists at its canonical path, emit a warning **before** the confirmation prompt:
   ```
   WARNING: Overwriting existing file: <path>
   ```
3. Prompt the user:
   ```
   Write these files? [y/N]
   ```
   Default is **N**.
4. If the user responds with anything other than `y`, exit cleanly:
   ```
   Export cancelled — no files written
   ```
   No files are written.
5. If the user responds `y`, proceed to Step 4.

These transformation prompts are self-instructions executed by the agent using its own in-context LLM reasoning. No external API call, subprocess, or tool invocation is required to apply them — the agent reads the prompt and generates the output directly.

#### Shared STRIP Preamble

The following items MUST be stripped from ALL target outputs. Each transformation prompt below references this block — apply it in addition to any target-specific delta listed in that prompt.

- All slash commands used as executable triggers (any `/<word>` pattern that is a Claude Code meta-tool or SDD phase command)
- Task tool references and sub-agent delegation patterns (`"Task tool:"`, `"subagent_type:"`, `"Launch sub-agent"`, `"Sub-agent launch pattern"`)
- install.sh and sync.sh references
- Claude Code-specific identity statements ("I am an expert development assistant…")
- The `## Skills Registry` section of `CLAUDE.md` (lines beginning with `~/.claude/skills/` or `.claude/skills/`)
- Any section whose content is enclosed between `<!-- [auto-updated]` and `<!-- [/auto-updated] -->` comment markers in `ai-context/` files

---

#### Copilot transformation prompt

Apply the following prompt to the source bundle to generate `.github/copilot-instructions.md`:

---

You are transforming a Claude Code project configuration into a GitHub Copilot instruction file.

**Source bundle:** CLAUDE.md + any available ai-context/ files provided above.

**STRIP:**

Apply the Shared STRIP Preamble above, then additionally strip:

- Plan Mode rules section (Claude Code-specific)

**ADAPT — do NOT strip; rewrite for Copilot:**

1. **SDD phase DAG and workflow** → Convert to a `## SDD Development Workflow` section written in declarative prose:
   - Describe each phase (explore, propose, spec, design, tasks, apply, verify, archive) and its artifacts
   - Include the artifact types (proposal, spec, design, tasks, verify-report, archive-report) and explain they are persisted to engram
   - Do NOT include any slash command syntax — describe the phases as steps the developer initiates

2. **SDD available commands table** → Replace with a `## Active SDD Coaching Instructions` section containing Copilot behavioral instructions:

   ```
   When a developer mentions implementing a new feature, change, or fix:
   - Proactively ask: "Would you like to follow the SDD workflow for this change?"
   - If yes, guide them step by step: propose → (spec + design in parallel) → tasks → apply → verify → archive
   - Before starting apply, confirm that proposal, design, and tasks artifacts exist in the project's engram
   - Remind the developer to run verify after implementation and archive once confirmed
   ```

   Adapt the phrasing to be Copilot-idiomatic (imperative instructions telling Copilot how to behave).

3. **Project memory layer (ai-context/)** → Retain the table and description verbatim; it applies directly to projects using Copilot too.

**RETAIN and adapt:**

- Tech stack (language, framework, key tools, versions)
- Coding conventions (naming, style, patterns) — rephrase as direct instructions to the AI assistant in imperative voice
- Architecture decisions and rationale
- Known issues and gotchas relevant to a developer working in the project
- Key working principles (clean code, no over-engineering, tests as first-class citizens, etc.)
- SDD artifact types and engram topic key patterns

**FORMAT:**

- Single flat Markdown file
- UTF-8, no BOM
- H2 sections (no YAML frontmatter)
- Start with a top-level