The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes dashboard data retrieved from the Statsig Console API, which represents an untrusted data ingestion surface.
Ingestion points: The output of scripts/read_dashboard.py containing data from the Statsig Console API.
Boundary markers: None identified in the prompt instructions to delineate external content from system instructions.
Capability inventory: The agent has the capability to write local files and execute network-enabled scripts (create_dashboard.py, add_dashboard_widgets.py, replace_dashboard_widgets.py).
Sanitization: No explicit sanitization or validation of the retrieved API data is performed before it is presented to the agent for summarization or transformation.
[SAFE]: Secret management follows best practices by using environment variables (STATSIG_CONSOLE_API_KEY) rather than hardcoding credentials.
[SAFE]: Network operations are directed to the official API domain for the tool's vendor.

statsig-dashboard