The Agent Skills Directory

[SAFE]: No security issues were detected. The skill is limited to performing business analytics queries and generating reports.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Amplitude feedback and experiments. Ingestion points: get_feedback_insights and get_experiments (SKILL.md). Boundary markers: Not present, but logic focuses on aggregate trends. Capability inventory: Restricted to Amplitude data retrieval and report generation; no shell, file-write, or arbitrary network capabilities. Sanitization: None. This surface is considered safe due to the restrictive toolset.
[DATA_EXPOSURE]: The skill accesses Amplitude project data using standard tool calls. No sensitive local environment files or hardcoded credentials were identified.

weekly-brief