skills/ilamanov/skills/ticket/Gen Agent Trust Hub

ticket

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes free-form text from users to generate structured tickets, introducing a standard indirect prompt injection surface.
  • Ingestion points: Untrusted data enters the agent context via the user's free-form description in Step 1 of SKILL.md.
  • Boundary markers: No explicit input validation or boundary delimiters are applied during the initial ingestion, though Step 5 specifies putting the raw text inside a fenced quote block for the ticket comment.
  • Capability inventory: The skill possesses the capability to query workspace metadata and create resources using the Linear MCP toolset.
  • Sanitization: Input validation is absent; however, the risk is heavily mitigated by Step 4, which establishes a strict human-in-the-loop checkpoint requiring explicit user approval before any write operations occur.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:39 AM
Security Audit — agent-trust-hub — ticket