The Agent Skills Directory

[SAFE]: The skill's primary function is to analyze Amplitude analytics data using provided platform tools. The instructions follow best practices for data pacing and day-over-day comparisons.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user feedback sources which could theoretically contain instructions intended to bias the briefing.
Ingestion points: Phase 2, step 5 utilizes get_feedback_insights to retrieve user-submitted feedback themes from Amplitude.
Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from following embedded prompts within feedback content.
Capability inventory: The agent's tools are restricted to Amplitude analytics queries (searching, metrics, experiments, and deployments) and do not include system-level access or arbitrary internet browsing.
Sanitization: No instructions for sanitizing or escaping the content of feedback insights are present.
[SAFE]: All referenced domains (amplitude.com) and tools are associated with the skill's official author. No obfuscated code, hidden URLs, or unauthorized network operations were found.

daily-brief