The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process an entire codebase, which constitutes untrusted data that could contain malicious instructions designed to subvert the analysis.\n
Ingestion points: SKILL.md directs the agent to "Use the entire codebase as the context for this skill."\n
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat content within the codebase as data rather than instructions, which could lead to the agent following instructions found in code comments.\n
Capability inventory: The skill is capable of writing HTML report files to the local file system at codepulse-mfa/reports/.\n
Sanitization: The skill lacks instructions for sanitizing or validating content retrieved from the codebase before it is interpolated into the final HTML report, potentially allowing for the injection of malicious scripts into the generated report if the template is not handled carefully.

codepulse-mfa