The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation and illustrative code snippets for a domain-specific language. It does not contain executable scripts, malicious instructions, or obfuscated content. All external links point to the author's official GitHub repository or well-known educational and industry sites.
[INDIRECT_PROMPT_INJECTION]: The skill describes a language compiler and runtime system, which inherently creates a surface for indirect prompt injection if the agent processes untrusted PCL source code.
Ingestion points: The documentation for PCLRuntime.executePersona and CodeGenerator in SKILL.md and references/CORE_CONCEPTS.md describes the processing of external source code and persona-directed input.
Boundary markers: No specific delimiters or instructions for the agent to ignore instructions embedded in the PCL code are present in the provided examples.
Capability inventory: The skill is configured with broad tool permissions including Execute, Write, and Debug, which could be targeted by untrusted code input.
Sanitization: No input validation or sanitization logic is demonstrated in the provided architectural overview.

pcl-expert