The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): No instructions designed to override agent behavior, extract system prompts, or bypass safety filters were identified.
[DATA_EXFILTRATION] (SAFE): No network operations (curl, wget, fetch) or access to sensitive local files (~/.ssh, .env) were found. The skill operates exclusively on geometric input provided by the user.
[REMOTE_CODE_EXECUTION] (SAFE): The skill uses uv run python to execute a local script. There are no patterns involving the download and execution of remote scripts or unverified third-party packages.
[OBFUSCATION] (SAFE): All content is in plain text. No Base64 encoding, zero-width characters, or homoglyph-based evasion techniques were detected.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill processes untrusted user input (WKT strings and coordinate lists) and passes them to a subprocess via command-line arguments.
Ingestion points: Arguments such as --coords, --geom, --g1, and --g2 in scripts/shapely_compute.py calls.
Boundary markers: Absent; inputs are passed directly as strings.
Capability inventory: Execution of a local Python script via uv run.
Sanitization: Not explicitly visible in the skill markdown; relies on the internal implementation of scripts/shapely_compute.py and the Shapely library's parser.

shapely-compute