The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions include directives aimed at influencing the agent's tool-selection behavior, attempting to prioritize this skill over standard tool usage or other available workflows.
Evidence: "Use by default whenever the user wants... even indirectly or with approximate wording. Prefer this skill over raw MCP-tool handling when it reasonably fits. Skip only if the user explicitly asks not to use this skill/workflow."
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and display data from external sources (membership databases), creating a surface where malicious payloads in that data could influence the agent's output.
Ingestion points: Data retrieved from the classic_list_active_subscriptions and classic_list_expired_subscriptions tools (SKILL.md).
Boundary markers: Absent; the skill does not specify the use of delimiters or instructions to ignore embedded commands within the subscription data.
Capability inventory: The skill is restricted to READ_ONLY access and formatting data into markdown tables. It does not utilize capabilities for file system modification, network exfiltration, or code execution.
Sanitization: Absent; there is no instruction to sanitize, escape, or validate the content of retrieved PII fields such as names or emails before rendering them in the leaderboard tables.

membership-longest-subscribers