The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional Markdown content and reference data without any executable scripts or binaries.
[DATA_EXPOSURE]: While the skill analyzes app metrics, it does not instruct the agent to access credentials or environment variables. It operates on data explicitly provided by the user or accessed via restricted tools within the session.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it ingests untrusted metrics data.
Ingestion points: Metric data provided by the user through AskUserQuestion or files accessed via Read, Glob, and Grep tools.
Boundary markers: Absent; the instructions do not define delimiters for user-provided data.
Capability inventory: The agent's capabilities are restricted to safe tools (Read, Glob, Grep, AskUserQuestion). No shell execution, dynamic code execution, or network capabilities are granted in the frontmatter or instructions.
Sanitization: No explicit sanitization or validation of the input metrics data is performed.
[PROMPT_INJECTION]: No evidence of jailbreak attempts, safety bypasses, or instructions to ignore previous system prompts was found in the skill text.

analytics-interpretation