The Agent Skills Directory

[SAFE]: The skill implements a legitimate interface for querying the OpenAlex scholarly database and does not contain any malicious logic, unauthorized system access, or persistence mechanisms.
[EXTERNAL_DOWNLOADS]: The skill documentation specifies the installation of the requests library via the uv package manager to facilitate communication with the OpenAlex API.
[DATA_EXFILTRATION]: The skill transmits a user-provided email address to https://api.openalex.org to access the API's 'polite pool' for higher rate limits, which is a standard and documented feature of the OpenAlex service for tracking usage and preventing abuse.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection. 1. Ingestion points: Academic paper metadata (titles, abstracts) fetched from the OpenAlex API in scripts/openalex_client.py and scripts/query_helpers.py. 2. Boundary markers: No explicit delimiters or warnings are used when presenting the fetched content to the agent. 3. Capability inventory: The skill uses requests.get for network communication and includes code examples for writing data to local CSV files. 4. Sanitization: No explicit validation or filtering of the fetched metadata is implemented in the provided scripts.

openalex-database