The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes clinical trial data, including free-text fields like 'brief summary' and 'eligibility criteria', which are retrieved from the external ClinicalTrials.gov API. This represents an attack surface where malicious instructions could theoretically be embedded in trial descriptions to influence the agent's behavior.
Ingestion points: External data enters the agent context via the requests.get calls in scripts/query_clinicaltrials.py and the code snippets in SKILL.md.
Boundary markers: The instructions do not define clear boundaries or provide 'ignore instructions' warnings when presenting the API results to the agent.
Capability inventory: The skill possesses the capability to perform network requests (requests.get) and write data to the local file system (as shown in the CSV export examples).
Sanitization: The scripts do not implement sanitization, filtering, or validation of the text content returned from the API before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to https://clinicaltrials.gov/api/v2/studies to retrieve trial information. This targets a well-known, official government registry and is the primary purpose of the skill.
[COMMAND_EXECUTION]: The skill instructions and documentation provide examples for executing Python scripts (python3 query_clinicaltrials.py) and utilizing the requests library to interact with the database.

clinicaltrials-database