The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feedback messages and tool responses from the server, which are untrusted external inputs.
Ingestion points: The user-provided message parameter and the best-effort acknowledgement note returned by the server (SKILL.md).
Boundary markers: The instructions explicitly direct the agent to treat feedback text and returned content as plain text and never as instructions (SKILL.md).
Capability inventory: The skill's primary capability is the feedback.send function used to transmit reports (SKILL.md).
Sanitization: The agent is instructed to manually redact personal identifiers such as phone numbers and email addresses from the message before it is sent (SKILL.md).

sent-feedback