The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted content from external files during its analysis workflow, which creates a surface for indirect prompt injection. This is an inherent risk for any analysis-oriented tool and is managed by the agent's runtime protections.
Ingestion points: Reads SKILL.md and auxiliary files from specified local directory paths.
Boundary markers: None explicitly implemented to delimit analyzed content from the agent's core instructions.
Capability inventory: Performs local filesystem read/write operations and executes a specific utility script (scripts/extract-yaml.sh).
Sanitization: Does not perform specific sanitization or filtering of the text being analyzed.
[COMMAND_EXECUTION]: The skill includes and executes a local shell script (scripts/extract-yaml.sh) to extract metadata from skill files. It also utilizes standard filesystem commands like ls to verify directory structures. These operations are restricted to the provided skill path and are necessary for the skill's primary function.

skill-quality-reviewer