The Agent Skills Directory

[Indirect Prompt Injection] (SAFE): The skill processes data from the Google Places API, which represents a potential surface for indirect prompt injection if an attacker were to control venue information.
Ingestion points: Venue names and descriptions enter the system via the search_places and get_place_details functions in src/local_places/google_places.py.
Boundary markers: The skill returns structured JSON to the agent; explicit prompt boundary markers are not implemented within the skill itself.
Capability inventory: The skill is limited to performing network requests to the Google API and does not have access to command execution or file system modifications.
Sanitization: The skill relies on Pydantic for data structure validation but does not sanitize string content for potential malicious instructions.
[Data Exposure & Exfiltration] (SAFE): The skill handles the Google API key through standard environment variable practices and only transmits it over HTTPS to the official Google API domain.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): Dependencies specified in pyproject.toml are standard, reputable packages from PyPI. No suspicious remote code execution patterns or unverified script downloads were detected.

local-places