The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted project files without defined safety boundaries.
Ingestion points: Uses Read, Grep, Glob, and Bash to scan project files like sitecore.json, .csproj, and appsettings.json.
Boundary markers: Absent; the instructions do not specify delimiters or provide warnings to ignore embedded instructions within scanned content.
Capability inventory: The skill allows access to Bash, Read, Grep, and Glob tools, which could be misused if the agent obeys instructions found inside local files.
Sanitization: Absent; there is no content filtering or validation performed on the data read from the project files.

cms-detect