The Agent Skills Directory

Indirect Prompt Injection Surface: The skill processes external, user-provided 'Test code' and 'Production code' as its primary inputs. While this is necessary for its function of analyzing assertion quality, it creates a potential surface where malicious instructions embedded in the analyzed code could influence the agent's behavior.
Ingestion Points: The 'Test code' and 'Production code' input fields defined in the SKILL.md file.
Capability Inventory: The skill performs analytical tasks such as reading, classifying, and computing metrics on code structures. It does not utilize high-risk capabilities like arbitrary command execution, network exfiltration, or persistence mechanisms.
Sanitization and Boundaries: The workflow does not explicitly describe the use of boundary markers or sanitization techniques to isolate the user-provided code from the analysis instructions. Implementing these would be a recommended practice to mitigate risks from untrusted data ingestion.

assertion-quality