The Agent Skills Directory

[EXTERNAL_DOWNLOADS] (LOW): The script communicates with https://clinicaltrials.gov/api/v2. While this is a reputable government service, it is not within the pre-defined whitelist of trusted domains.
[PROMPT_INJECTION] (LOW): The skill serves as an ingestion point for external data, creating an indirect prompt injection surface.
Ingestion points: Data is fetched from the ClinicalTrials.gov API in the search_studies and get_study_details functions within scripts/query_clinicaltrials.py.
Boundary markers: Absent. The script returns raw JSON/text data from the API directly to the agent context without delimiters or warnings.
Capability inventory: None. The script only performs read operations and does not possess capabilities for file writing, command execution (eval/exec/subprocess), or secondary network exfiltration.
Sanitization: Absent. External content is not sanitized or filtered before being returned.
[DATA_EXFILTRATION] (SAFE): The script does not access sensitive local files (e.g., SSH keys, AWS credentials) or environment variables. All network traffic is directed to the designated API endpoint for query purposes.
[REMOTE_CODE_EXECUTION] (SAFE): No patterns for remote code execution or dynamic evaluation of API-returned content were found.

clinicaltrials-database