distill-sessions
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes common Unix utilities such as
find,grep,stat,perl,jq, andawkto identify and parse local JSONL log files. These commands are used for their intended purpose of log rotation and data extraction without privilege escalation. - [DATA_EXPOSURE]: The skill is configured to access interaction logs located in
~/.claude/and~/.codex/. To mitigate potential data leakage, Step 5 of the instructions explicitly mandates the redaction of emails, API keys, tokens, and passwords from any evidence presented to the user. - [INDIRECT_PROMPT_INJECTION]: The skill processes historical session data which may contain untrusted content from previous interactions. The risk is minimized by the use of structured data processing (
jq) and a human-in-the-loop workflow where the agent only proposes changes for the user to manually apply.
Audit Metadata