The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface inherent to its function as a code analysis tool.
Ingestion points: The skill ingests untrusted source code data from external repositories using the Read, Grep, Glob, and ast-grep tools.
Boundary markers: No explicit boundary markers or 'ignore' instructions are provided to the agent to distinguish between its own logic and instructions that might be embedded in code comments or strings within the analyzed files.
Capability inventory: The skill has access to restricted Bash commands (sg, rg), file system modification via TodoWrite, and Task management.
Sanitization: Content from the analyzed files is processed directly by the model without sanitization or filtering for potentially malicious instructions.

code-antipatterns-analysis