cost-session

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides transparency into message costs without any detected malicious patterns, obfuscation, or unauthorized data transmission.
  • [DATA_EXPOSURE]: The skill accesses session logs in the local directory ~/.claude/projects/ to extract token usage information. This activity is strictly confined to the skill's stated purpose of cost analysis.
  • [COMMAND_EXECUTION]: The skill executes a local script (scripts/session.mjs) using Bash to process and format cost data. This is standard operational behavior for local development tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 12:35 PM
Security Audit — agent-trust-hub — cost-session