The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting and acting upon instructions found within Kata issues.
Ingestion points: Data from issue bodies and comments is read using kata show, kata search, and kata ready in SKILL.md.
Boundary markers: The skill lacks explicit instructions or delimiters to treat ingested issue content as untrusted.
Capability inventory: The agent is authorized to execute kata CLI commands and arbitrary validation scripts (test, lint, build) found within the project or tracker.
Sanitization: No sanitization or escaping is performed on content retrieved from the issue ledger before processing.
[COMMAND_EXECUTION]: The skill requires the execution of multiple local shell commands to manage project state and verification.
Evidence: SKILL.md provides instructions for running kata commands for issue management and specifies that the agent should discover and run repository-specific validation commands like test, lint, and build.

kata-goal-delivery