The Agent Skills Directory

[SAFE]: The skill focuses on marketing logic and keyword categorization. The provided JavaScript snippets implement standard data processing for keyword discovery and grouping without invoking dangerous system calls or external network requests to untrusted domains.
[INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface where it processes external search term reports from Google Ads.
Ingestion points: The mineNegativeKeywords and searchTermReportMining functions in SKILL.md ingest external search term data.
Boundary markers: None are present in the logic snippets, as they focus on internal classification.
Capability inventory: The skill performs intent classification, deduplication, and ad group grouping; no dangerous capabilities such as file writes, network exfiltration, or subprocess execution are present.
Sanitization: The logic uses toLowerCase() and signal-based substring matching for intent classification, which is appropriate for its stated purpose.

keyword-research