The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill framework involves reading and analyzing project files to inform architectural decisions, creating a surface where instructions embedded in those files could influence agent output.
Ingestion points: Project files, requirements, and system design documents accessed using the Read, Glob, and Grep tools as described in SKILL.md.
Boundary markers: The instructions lack explicit boundary markers or directions for the agent to ignore instructions contained within the data being analyzed.
Capability inventory: The skill is restricted to read-only tools (Read, Glob, Grep) via the allowed-tools frontmatter in SKILL.md. It has no permissions for network access, shell execution, or file modification.
Sanitization: There is no evidence of sanitization or filtering of the content read from external files.

architecture