Skip to content
Back to skills

Usgs Seismic Alert

ASecurity

Poll the USGS M4.5+ earthquake feed, conditionally pause for Incident Commander triage when magnitude >= 5.0, and emit an incident bulletin using examples/usgs_seismic_alert/actions.py.

  • 12 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 3, 2026
ai-agentspythonbash

Security analysis

A100/100

Scanned October 3, 2026

npx -y skills add google/lightflow --skill usgs_seismic_alert --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Usgs Seismic Alert?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Usgs Seismic Alert
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/google-usgs-seismic-alert/badge)](https://www.skillsdirectory.com/skills/google-usgs-seismic-alert)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: usgs-seismic-alert
description: Poll the USGS M4.5+ earthquake feed, conditionally pause for Incident Commander triage when magnitude >= 5.0, and emit an incident bulletin using examples/usgs_seismic_alert/actions.py.
---

# USGS Seismic Alert Triage (`usgs_seismic_alert`)

Executes the 3-stage USGS earthquake alert workflow using
`examples/usgs_seismic_alert/actions.py`.

## Workflow Stages & Rules

1.  **Stage 1 — Fetch USGS Feed (`fetch_usgs_feed`)**:

    -   Call `actions.fetch_usgs_feed(payload)` with `{"offline": True}` (or
        live GeoJSON feed if `offline=False`).
    -   Note: `fetch_usgs_feed` returns a single `dict` (not a tuple):
        `{"events": [...], "event_count": int, "max_magnitude": float,
        "strongest_place": str, "source": str}`.
    -   Persist this dict under `payload["outputs"]["fetch_usgs_feed"]` so Stage
        1 is **not** re-executed after triage.
    -   Example invocation:

        ```bash
        python3 -B -c "import json, sys; sys.path.insert(0, 'examples/usgs_seismic_alert'); import actions; out = actions.fetch_usgs_feed({'offline': True}); print(json.dumps({'outputs': {'fetch_usgs_feed': out}}))"
        ```

2.  **Stage 2 — Conditional Incident Commander Gate
    (`incident_commander_gate`)**:

    -   **Branch Condition (`run_if`)**: Evaluate
        `outputs.fetch_usgs_feed.max_magnitude >= 5.0`.
    -   If `max_magnitude < 5.0`, skip Stages 2 and 3.
    -   If `max_magnitude >= 5.0`, **STOP** and present `max_magnitude` and
        `strongest_place` to the human Incident Commander. Wait for their
        explicit approval and `severity` (`"ADVISORY" | "WATCH" | "WARNING"`) +
        `output_path`. Never auto-approve.

3.  **Stage 3 — Emit Bulletin (`emit_bulletin`)**:

    -   **Condition**: Run only if Stage 2 was approved. Do **not** re-call
        `fetch_usgs_feed`.
    -   Call `actions.emit_bulletin(payload)` where `payload` contains:
        -   `"outputs": {"fetch_usgs_feed": <stage_1_out>,
            "incident_commander_gate": {"severity": "<severity>", "output_path":
            "<path>"}}`
    -   Returns `({"bulletin_path": str, "severity": str, "simulated": bool},
        msg)`.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.