handing-off

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTION

Full Analysis

[COMMAND_EXECUTION]: The skill uses shell scripts (mark.sh and check.sh) to manage session metadata. These scripts interact with standard directories (e.g., ~/.claude/) to store state files and read transcripts. The commands use standard utilities like find, grep, sed, and date to process local files and calculate token usage for context management.

Audit Metadata

Risk Level

SAFE

Analyzed

Sep 3, 2026, 06:02 PM

Security Audit — agent-trust-hub — handing-off