ulw-statuslisted
Install: claude install-skill X0x888/oh-my-claude
# ULW Status
Parse any argument passed with the skill invocation (stored in `$ARGUMENTS`).
Treat the arguments case-insensitively.
- If the user passed `summary`, `-s`, or `--summary`, run the compact recap:
```
bash ~/.claude/skills/autowork/scripts/show-status.sh --summary
```
- If the user passed `classifier`, `-c`, or `--classifier`, run the
intent-classifier telemetry surface:
```
bash ~/.claude/skills/autowork/scripts/show-status.sh --classifier
```
- If the user passed `explain`, `-e`, or `--explain`, run the per-flag
rationale (v1.30.0):
```
bash ~/.claude/skills/autowork/scripts/show-status.sh --explain
```
- Otherwise, run the full diagnostic status:
```
bash ~/.claude/skills/autowork/scripts/show-status.sh
```
Display the output as-is.
The `--summary` form is the right choice at end-of-session — it folds session
duration, edit counts, guard blocks, classifier misfires, reviewer verdicts,
and commits made during the session onto ~8 lines. It surfaces "invisible
friction" signals that the full dump buries, so the user can see patterns
(e.g. "3 sessions in a row with mostly-misfire PreTool blocks → tune the
classifier") that would otherwise live only in memory.
The `--explain` form (v1.30.0) walks the active `oh-my-claude.conf` and
prints each flag's current value, default, and one-line purpose, grouped
by category. Non-default values are flagged with a `*` marker. Use this
when you want to disable a flag (e.g., `intent_broadeni