The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, potentially untrusted data to perform audits. This data could contain malicious instructions designed to manipulate the audit outcome or influence the agent's behavior.
Ingestion points: Data enters the context via ai_system_description, use_case, and existing_documentation variables defined in SKILL.md.
Boundary markers: Absent. The instructions do not specify delimiters or provide guidance to the agent to ignore instructions embedded within the audit evidence.
Capability inventory: The skill utilizes an ExecutionProxy Interface for terminal command execution and a MemoryStore Interface for writing persistent markdown files to local directories (e.g., Obsidian docs/).
Sanitization: Absent. No explicit filtering or sanitization steps are defined for the input data.
[COMMAND_EXECUTION]: The skill's global protocols define the use of an ExecutionProxy Interface (prefixed with rtk) for executing terminal actions such as rtk npm test. This grants the agent the ability to execute shell commands within the environment, though it is framed within a TDD (Test-Driven Development) methodology.

iso-42001-ai-governance