The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves processing external test datasets, which creates a surface for indirect prompt injection attacks.
Ingestion points: Untrusted data enters the agent context via tests.json dataset files referenced in SKILL.md and references/templates.md.
Boundary markers: There are no explicit boundary markers or "ignore embedded instructions" warnings present in the skill to delimit test data from prompt instructions.
Capability inventory: The skill is configured to use Bash, Read, and Write tools, which provides a significant capability surface if an injection were to occur.
Sanitization: The skill does not define any sanitization, escaping, or validation protocols for external test case content before it is processed by the agent.

prompt-testing