The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill relies on external Python packages and interacts with public security services.
Utilizes the pycrtsh library to query Certificate Transparency databases.
Uses the requests library to fetch JSON data from https://crt.sh.
References certstream (wss://certstream.calidog.io/) for real-time certificate monitoring.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Certificate Transparency logs. Since certificate fields (like Common Name or SANs) can be populated by any entity issuing a certificate, they could contain malicious instructions designed to influence an AI agent.
Ingestion points: scripts/agent.py retrieves records from the crt.sh API and via the pycrtsh library.
Boundary markers: The skill does not use specific delimiters or instructions to treat the retrieved certificate data as untrusted content.
Capability inventory: The script performs file operations (writing ct_report.json) and network requests to established certificate databases. It does not execute dynamic code based on the input.
Sanitization: There is no evidence of sanitization or filtering of the certificate strings before they are printed to the console or saved to the output report.

analyzing-tls-certificate-transparency-logs