The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides specific shell commands for the agent to use during discovery to explore the local project filesystem.- [COMMAND_EXECUTION]: The agent is explicitly directed to read sensitive files, including all environment files (.env) and secrets management configurations, to understand system assumptions. This over-privileged access to credentials and tokens presents a high risk of data exposure.- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by design, as its core functions involve automated ingestion and interpretation of content from various codebase files.- [PROMPT_INJECTION]: Evidence Chain for Indirect Injection:- Ingestion points: Source code (src/), database migrations, environment files, and infrastructure configurations.- Boundary markers: No explicit delimiters or instructions are provided to distinguish between the agent's instructions and the content being read from files.- Capability inventory: The agent can execute filesystem discovery commands and is encouraged to pass information to other agents like the Reality Checker or Backend Architect, potentially propagating malicious instructions.- Sanitization: No sanitization, escaping, or validation of the file content is performed before the agent processes it.

agency-workflow-architect