The Agent Skills Directory

[Prompt Injection] (HIGH): The skill is highly susceptible to indirect prompt injection due to its broad file-access permissions and write capabilities.\n
Ingestion points: The agent uses Read, Glob, and Grep to ingest content from Rails models, controllers, and configuration files.\n
Boundary markers: None. There are no instructions or delimiters defined to prevent the agent from following malicious instructions embedded within code comments or string literals in the files being analyzed.\n
Capability inventory: The skill utilizes Write, Edit, and Bash tools, allowing it to modify the codebase or execute system-level commands.\n
Sanitization: None. The skill performs no validation on the code it analyzes before using it to inform its actions.\n- [Command Execution] (MEDIUM): The skill is granted use of the Bash tool for environment setup and management tasks. While standard for Rails development, this capability can be exploited if the agent's logic is subverted through the aforementioned injection vulnerabilities.

performance-optimization