The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface.
Ingestion points: Untrusted ticket data ingested through FRESHSERVICE_LIST_TICKETS and FRESHSERVICE_GET_TICKET as documented in SKILL.md.
Boundary markers: The instructions lack explicit delimitation or warnings to the agent to ignore instructions embedded within retrieved ticket descriptions.
Capability inventory: The skill provides extensive write capabilities including FRESHSERVICE_CREATE_TICKET, FRESHSERVICE_BULK_UPDATE_TICKETS, and FRESHSERVICE_CREATE_TICKET_OUTBOUND_EMAIL which could be manipulated by malicious ticket content.
Sanitization: No evidence of input validation or content sanitization before processing ticket data.
[EXTERNAL_DOWNLOADS]: External Infrastructure Dependency.
The skill documentation in SKILL.md directs users to connect to https://rube.app/mcp for tool functionality. This is a functional dependency for the Rube/Composio MCP integration.
[SAFE]: Self-referential Security Metadata.
The skill-report.json file contains internal audit data (security_audit block) that asserts a 'safe' risk level and 'safe_to_publish' status. These claims are evaluated as metadata and do not influence the independent security verdict.

freshservice-automation