The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions include standard development setup steps that invoke external package managers like 'npm' and 'pip' to download dependencies.
[COMMAND_EXECUTION]: The skill provides commands to execute local Python scripts intended for project analysis and the application of automated fixes to the filesystem.
[PROMPT_INJECTION]: The skill's core functionality of analyzing external codebases creates an attack surface for indirect prompt injection.
Ingestion points: The skill ingests untrusted code data from the user's filesystem via parameters such as <target-path> and <project-path> used in its scripts.
Boundary markers: Absent; there are no instructions to use delimiters or ignore potentially malicious commands embedded within the analyzed project files.
Capability inventory: The skill utilizes Python scripts capable of reading filesystem content and applying 'automated fixes' (write access).
Sanitization: Absent; no evidence of input validation or content escaping for the data extracted from the analyzed projects is provided.
[NO_CODE]: The skill references multiple implementation scripts in the scripts/ directory that were not included in the provided file package.

senior-architect