The Agent Skills Directory

[SAFE]: The skill implements a local health tracking system that stores data in data/health-goals-tracker.json. It operates within a limited scope, using only the Read and Write tools for its stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input, such as goal descriptions and habit names, which creates a potential surface for indirect prompt injection.
Ingestion points: User-provided strings for goal setting and habit recording in SKILL.md and examples.md.
Boundary markers: Absent; there are no specific instructions for the agent to treat the input as data or to ignore embedded instructions.
Capability inventory: Read and Write tools are used to manage local JSON files.
Sanitization: Absent; no explicit filtering or sanitization of input strings was found.
However, the risk is low as the skill lacks network access or high-privilege capabilities to exploit such an injection.

goal