pixie-status

name	pixie-status
description	Shows Pixie's current runtime state - whether the server is up, which tools are warm, ports, recent activity, uptime - read-only. Use when the user asks if Pixie is running, what's running now, or what's warm. Do NOT use if Pixie is broken (pixie-doctor) or to enumerate installed tools (list-tools).
allowed-tools	Bash, Read, Glob

Pixie runtime snapshot

You are producing a one-screen snapshot of the running Pixie process: uptime, currently-warm tool subprocesses, recent run count, database size. This is a live-state question, distinct from the installation-health check in pixie-doctor.

Routing check (do this first)

• If the user wants a static install health check (Python version, deps), switch to pixie-doctor.
• If the user wants the list of installed tools with cached validation, switch to list-tools.
• If the user wants live stderr from one running tool, switch to view-logs.

Pre-flight: read the kill file

Before doing anything that might touch existing patterns, read .build/KILL_FILE.md. Any entry whose Context or Symptom matches your current task -> apply the documented Fix directly, do not re-debug from scratch.

Steps

1. Determine the API port

Default 7860. Honour PIXIE_PORT if set.

PowerShell: $env:PIXIE_PORT Bash: echo "${PIXIE_PORT:-7860}"

2. Call /api/healthz

uv run python -c "
import httpx, os, json
port = os.environ.get('PIXIE_PORT', '7860')
try:
    r = httpx.get(f'http://127.0.0.1:{port}/api/healthz', timeout=2.0)
    r.raise_for_status()
    print(json.dumps(r.json(), indent=2))
except httpx.ConnectError:
    print('PIXIE_NOT_RUNNING')
except Exception as e:
    print('ERROR: ' + str(e))
"

Note for the orchestrator: this assumes GET /api/healthz returns a JSON document with at least: uptime_seconds, warm_tools (list of {id, pid, port, started_at, rss_mb?}), db_size_bytes, recent_runs_24h. If the endpoint isn't shipped yet, this skill returns "not running" — the same as if the process were down.

3. Branch on the response

Case A — PIXIE_NOT_RUNNING

Tell the user plainly:

"Pixie is not running on port <port>. Start it with:

uv run pixie

from the repo root (C:\dev\Pixie)."

Stop.

Case B — Pixie responded

Render a markdown summary:

## Pixie runtime snapshot

- **Uptime:** <human-readable, e.g., "2h 17m">
- **Listening on:** 127.0.0.1:<port>
- **Warm tools:** N
- **Recent runs (24h):** M
- **Database size:** <human-readable, e.g., "3.4 MiB">

Then a warm-tools table (only if N > 0):

Tool	PID	Port	Warm since	RSS

Warm since is the relative time (e.g., "12m ago"). RSS is in MiB if reported, else —.

If N == 0:

"No tools are currently warm. Open a tool in the dashboard to spin one up."

4. Footer

End with a single line:

"Snapshot at . Re-run the skill to refresh."

On failure: append to the kill file

If you encounter an error NOT already in .build/KILL_FILE.md, append a new entry with the next KILL-NNNN id following the schema. Be terse -- root cause + one-line fix + one-line rule.

Do NOT

• Do NOT start, stop, or restart Pixie.
• Do NOT call any non-GET endpoint.
• Do NOT modify any file.
• Do NOT run the validator.
• Do NOT bind to 0.0.0.0 or any non-loopback interface.
• Do NOT invoke other Pixie skills programmatically. Offer the user a choice and stop.