The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the addition of an external MCP server endpoint located at https://rube.app/mcp as part of the client configuration to access the PagerDuty toolset.
[PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection by ingesting and processing data from external PagerDuty incident reports and alerts.
Ingestion points: The skill retrieves external content via tools such as PAGERDUTY_FETCH_INCIDENT_LIST, PAGERDUTY_GET_ALERTS_BY_INCIDENT_ID, and PAGERDUTY_RETRIEVE_SERVICE_BY_ID.
Boundary markers: There are no specific boundary markers or instructions to isolate the retrieved incident titles or alert details from the agent's core instruction set.
Capability inventory: The skill possesses state-changing capabilities, including PAGERDUTY_UPDATE_INCIDENT_BY_ID, PAGERDUTY_CREATE_INCIDENT_RECORD, and PAGERDUTY_CREATE_NEW_SERVICE, which could be exploited if malicious instructions are present in the ingested data.
Sanitization: No data sanitization or validation protocols are described to filter potentially harmful content from the PagerDuty API responses before they enter the agent's context.

pagerduty-automation