Run AI tasks through the low-latency fast client path.
Copy the command definition below into:
~/.claude/commands/fast.md# f fast
Run AI tasks through the low-latency fast client path.
This command is optimized for hot-loop invocation and prefers `fai`/`ai-taskd-client` over full `f` task startup.
## Usage
```bash
f fast ai:flow/noop
f fast ai:flow/bench-cli -- --iterations 30
f fast --root ~/code/flow ai:flow/dev-check
f fast --no-cache ai:flow/dev-check
```
## Behavior
1. Tries fast client dispatch (`fai`, local `target/.../ai-taskd-client`, or `ai-taskd-client` on PATH).
2. If daemon is not running, starts `ai-taskd` and retries.
3. Falls back to direct daemon dispatch if no fast client binary is found.
## Options
- `--root <PATH>`: root directory for `.ai/tasks` discovery (default `.`)
- `--no-cache`: disable cached binary execution and use direct Moon run
- `TASK`: required AI selector like `ai:flow/dev-check`
- trailing args after `--` are passed to the task
## Notes
- `f fast` is intentionally for `ai:*` selectors only.
- Install `fai` for best latency:
```bash
f install-ai-fast-client
f tasks daemon start
f fast ai:flow/noop
```
For pooled burst execution and timings, use `fai` directly:
```bash
fai --timings ai:flow/noop
printf 'ai:flow/noop\nai:flow/noop\n' | fai --batch-stdin --timings
```