The Agent Skills Directory

[SAFE]: The skill's primary function is to wrap existing platform tools (hooks_route) to provide cost-efficiency reports. No unauthorized data access or malicious intent was detected.
[COMMAND_EXECUTION]: Employs the Bash tool to perform routine operations such as timestamp generation and interacting with a local memory store. These operations are limited to the skill's stated purpose of tracking and reporting.
[DATA_EXPOSURE]: Interacts with memory_search and memory_list to aggregate task history for auditing. This data access is localized to the agent's internal memory and does not involve exfiltration to external or unauthorized services.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes task descriptions which may contain untrusted content.
Ingestion points: Task descriptions are ingested via CLI arguments or the memory_search tool (Step 1).
Boundary markers: None present to distinguish instructions from task data.
Capability inventory: Uses the Bash tool to store data and mcp__claude-flow__hooks_route for classification.
Sanitization: No explicit sanitization or escaping of the ingested task descriptions is performed before they are processed by the router.

cost-booster-route