The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documents and utilizes numerous reputable OSINT and registry services for data retrieval, including GLEIF for LEI records, SEC-EDGAR for corporate filings, OpenCorporates for entity verification, and RIR databases (ARIN/RIPE) for network block attribution. These are well-known, authoritative sources aligned with the skill's discovery purpose.
[COMMAND_EXECUTION]: The instructions provide explicit curl and PowerShell templates for interacting with API endpoints. These commands are transparently documented for the agent to use or present to the user and are scoped to data retrieval from the mentioned reputable sources.
[SAFE]: The skill implements significant security guardrails, notably the 'HYPERSCALER-SCOPE GUARD,' which prevents the misattribution of entire shared cloud provider ranges to a single tenant. It also utilizes a structural 'discover-only' flag to ensure that any newly discovered assets are treated as leads and are never automatically promoted to active scanning without explicit human intervention.
[DATA_EXFILTRATION]: No malicious data exfiltration patterns were detected. The skill focuses on gathering public OSINT data and explicitly instructs the agent to avoid providing sensitive PII or client details to cloud models unnecessarily.

org-attack-surface