The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which may contain embedded instructions designed to influence agent behavior.
Ingestion points: The feature description provided via $ARGUMENTS and the analysis of the existing codebase in Step 4.
Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or triple quotes) or explicit warnings to the agent to ignore instructions within the ingested data.
Capability inventory: The skill utilizes several powerful tools including Bash, Edit, and Read, along with task management and memory storage tools from the mcp__plugin_ruflo-core_ruflo suite.
Sanitization: No sanitization, validation, or escaping of the input data is performed before it is analyzed or stored in memory.

sparc-spec