epic-requirements-specification

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill interactively ingests user input to build an Epic specification which is eventually persisted to the file system and potentially synced with external tracking systems via MCP. This creates a surface for indirect prompt injection where instructions embedded in project requirements could influence downstream processes.
  • Ingestion points: User responses during the Phase 1 Interactive Discovery process in SKILL.md.
  • Boundary markers: The skill does not explicitly define sanitization or delimiters for the user-supplied content in the generated Markdown structure.
  • Capability inventory: The skill has the capability to write Markdown files to the workspace (epic-<slug>.md) and trigger follow-up actions like syncing to tracking systems or invoking the write-user-story skill (Phase 3).
  • Sanitization: No explicit sanitization of user input is documented. However, the design includes a mandatory "Approval Gate" in Phase 2, requiring the user to explicitly review the draft and type "APPROVED" before any automated execution occurs, providing a significant human review checkpoint.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 09:52 PM
Security Audit — agent-trust-hub — epic-requirements-specification