The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/plot_orbits.py executes local helper scripts (manage_ground_stations.py and fetch_tle.py) to resolve satellite and station data. These calls use subprocess.run() with argument lists rather than shell strings, effectively mitigating shell injection risks from user-supplied inputs such as spacecraft names or station identifiers.
[EXTERNAL_DOWNLOADS]: The skill leverages the uv tool to manage Python dependencies declared via PEP 723 inline metadata. This involves downloading established scientific libraries including skyfield, matplotlib, and numpy from official package registries at runtime.
[PROMPT_INJECTION]: The skill ingests untrusted data including spacecraft names, ground station identifiers, and Two-Line Element (TLE) strings. While this represents an attack surface for indirect prompt injection, the risk is minimized as the data is consumed by scientific libraries that do not interpret the content as executable instructions.
Ingestion points: CLI arguments parsed in scripts/plot_orbits.py and JSON data returned from local subprocesses.
Boundary markers: Absent; data is passed directly to library constructors and functions.
Capability inventory: Writing PNG files to a local directory and executing specific local Python scripts via uv run.
Sanitization: Filenames for the generated plots are sanitized using an alphanumeric filter in _safe_filename() to prevent path traversal or other malicious filesystem operations.

ops-orbit-plotter