The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection.\n
Ingestion points: The skill reads task data from the ~/todos.md file.\n
Boundary markers: No delimiters or explicit instructions to ignore embedded commands are defined when parsing task content.\n
Capability inventory: The skill utilizes Read and Write tools to modify the local file system.\n
Sanitization: No input validation or sanitization is performed on task text before it is displayed or processed.\n- [SAFE]: No evidence of unauthorized network activity or data exfiltration was detected.\n- [SAFE]: The skill does not download external dependencies or execute remote code.

task-tracker