skills/ikkeseb/skills/handoff/Gen Agent Trust Hub

handoff

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONTIME_DELAYED_CONDITIONAL
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill aggregates session history, goals, and plans to create a handoff file for future sessions. This creates a surface where malicious instructions present in the current conversation could be persisted into the markdown handoff and subsequently interpreted as instructions by an agent in a new session.
  • Ingestion points: Session memory and conversation history (referenced in SKILL.md).
  • Boundary markers: Absent; the skill does not specify wrappers or delimiters to prevent the agent from obeying instructions embedded in the summarized content.
  • Capability inventory: File writing, directory creation, and file deletion (SKILL.md).
  • Sanitization: Absent; the instructions prioritize matching user language over escaping potentially malicious input.
  • [COMMAND_EXECUTION]: The skill requires the agent to perform file system operations including directory creation (mkdir) and file deletion (rm) within the user's configuration directories (e.g., $HOME/.claude/handoffs). While these are intended for task management, they involve direct manipulation of the host file system.
  • [TIME_DELAYED_CONDITIONAL]: The skill includes logic to automatically delete files with a modification time older than 30 days. While scoped to the specific handoff directory, this represents a conditional file deletion capability triggered by time.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 09:27 AM