# Ask

> Send the user's request to specified AI provider asynchronously

- Skill: `majiayu000/ask-4` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/ask-4`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/ask-4/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/ask-4

---


# Ask AI Provider

Send the user's request to the specified AI provider via ask.

## Usage

The first argument must be the provider name. The message MUST be provided via stdin
(heredoc or pipe), not as CLI arguments, to avoid shell globbing issues:
- `gemini` - Send to Gemini
- `claude` - Send to Claude
- `opencode` - Send to OpenCode
- `droid` - Send to Droid

## Execution (MANDATORY)

```bash
CCB_CALLER=codex ask $PROVIDER <<'EOF'
$MESSAGE
EOF
```

## Rules

- After running the command, say "[Provider] processing..." and immediately end your turn.
- Do not wait for results or check status in the same turn.
- The task ID and log file path will be displayed for tracking.

## Examples

- `/ask gemini What is 12+12?` (send via heredoc)
- `CCB_CALLER=codex ask gemini <<'EOF'`
  `What is 12+12?`
  `EOF`

## Notes

- If it fails, check backend health with the corresponding ping command (`ping <provider>` (e.g., `ping gemini`)).
- Codex-managed sessions default to foreground; use `--background` or `CCB_ASK_BACKGROUND=1` for async.