The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process and apply content from external files (GLOSSARY.md and GLOSSARY-MAP.md). This establishes a surface for indirect prompt injection where instructions embedded in these data files could influence agent behavior.
Ingestion points: GLOSSARY.md and GLOSSARY-MAP.md are referenced as sources of terminology in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within these files.
Capability inventory: The skill explicitly sets disable-model-invocation: true in the frontmatter of SKILL.md, preventing the agent from executing tools or scripts.
Sanitization: No sanitization or validation of the glossary file content is mentioned.

wait-what