The Agent Skills Directory

Indirect Prompt Injection Surface: The skill is designed to process and act upon user feedback comments. While this is the intended purpose, it involves the ingestion of untrusted data from the user.
Ingestion points: User feedback is retrieved using the listComments tool as described in step 1 of the skill instructions.
Boundary markers: The instructions do not explicitly specify delimiters or boundary markers for the comment content when processed by the agent.
Capability inventory: The agent is authorized to make changes to the codebase and manage comment status using the resolveComments and deleteComments tools.
Sanitization: No explicit sanitization or validation of the comment content is mentioned before the agent acts on the feedback.

act-on-feedback