session-handoff

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a formatting and process guide for session management. It contains specific safety directives (Step 9 and Compression Rules) requiring the AI to sanitize output by removing tokens, passwords, and sensitive URLs before finalizing the handoff note.- [DATA_ACCESS]: The skill provides instructions for the agent to query its own local session database (sqlite_master) to recover lost drafts. This is an operational feature for context recovery and does not involve unauthorized access or external exfiltration.- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data from previous sessions.
  • Ingestion points: Pasted handoff notes from other sessions (SKILL.md).
  • Boundary markers: Instructions mandate the use of fenced markdown blocks and a self-identifying first line to distinguish handoffs from instructions (Workflow Step 8).
  • Capability inventory: File reading, terminal command execution, and local database querying (SKILL.md).
  • Sanitization: Explicit workflow steps (Step 9) to review and remove secrets, private account identifiers, and sensitive URLs.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 08:22 AM
Security Audit — agent-trust-hub — session-handoff