Skip to the content.

AI Agent Configuration

This repo keeps a single canonical prompt in PROMPT_SYSTEM.md and all agent files reference it using a relative path.

Files and minimal contents

GEMINI.md (repo root or subdir)

# shellgAIde
@PROMPT_SYSTEM.md

CLAUDE.md (repo root or subdir)

# shellgAIde
@PROMPT_SYSTEM.md

AGENTS.md (repo root, for Codex/Copilot)

# shellgAIde - GNU-first Shell Toolchain
See full system prompt: [PROMPT_SYSTEM.md](PROMPT_SYSTEM.md)

Core rules: Google Shell Style Guide, GNU tools (grep/sed/date), Bash 5+,
main() structure, CI-safe.

.github/copilot-instructions.md (subdir)

# shellgAIde Copilot
Follow [../PROMPT_SYSTEM.md](../PROMPT_SYSTEM.md): GNU-first toolchain, Google
Shell Style, main() structure.

Notes on paths

Manual setup (full instructions)

1) ChatGPT

Recommended approach:

  1. Create a ChatGPT Project (or use Custom Instructions) and paste the full content of PROMPT_SYSTEM.md into the project instructions.
  2. Ask for either:
    • a new script (“Create a new script that does X”), or
    • a refactor (“Refactor this script to conform to the system prompt; preserve behavior”).

When refactoring, include:

2) OpenAI Codex (CLI)

Codex CLI supports repository instructions via AGENTS.md. Typical flow:

  1. Start Codex in the repo root.
  2. Run /init to create AGENTS.md.
  3. Paste the full content of PROMPT_SYSTEM.md into AGENTS.md (or summarize while keeping the non-negotiable rules).
  4. Ask Codex to create/refactor the script, then review and apply diffs.

2b) OpenAI Codex (VS Code extension)

The VS Code Codex extension also supports repository instructions. Typical flow:

  1. Ensure AGENTS.md exists in the repo root.
  2. Paste the full content of PROMPT_SYSTEM.md into AGENTS.md (or a concise but complete adaptation).
  3. Reload the VS Code window or restart the Codex session.
  4. Use Codex Chat to create/refactor scripts; the instructions are applied automatically.

3) GitHub Copilot (VS Code / Visual Studio / JetBrains)

Copilot supports repository custom instructions. Typical flow:

  1. Create .github/copilot-instructions.md in your target repo.
  2. Paste the full content of PROMPT_SYSTEM.md into that file (or a concise but complete adaptation).
  3. Use Copilot Chat to create/refactor scripts with the instructions applied automatically.

Note: Copilot instruction support may vary by IDE and requires the feature to be enabled.

4) Google Gemini (Gemini CLI)

Recommended approach:

  1. Add a GEMINI.md file in your repo root (or parent dirs).
  2. Reference PROMPT_SYSTEM.md with a relative @PROMPT_SYSTEM.md.

Notes:

5) Anthropic Claude (Claude Code)

Recommended approach:

  1. Add a CLAUDE.md file in your repo root (or parent dirs).
  2. Reference PROMPT_SYSTEM.md with a relative @PROMPT_SYSTEM.md.
  3. Optionally run /init in Claude Code to generate a starter CLAUDE.md, then review and refine it to match your real workflow.

Notes:

Verify output with this toolchain

After generating/refactoring a script (locally):

make setup
# then inside the toolchain shell:
make lint
make test
make ci

For a single script:

make shell
# then:
./bin/lint.sh path/to/script.sh