session-logs
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs common Unix utilities such as
jq,rg(ripgrep),awk,sort, andgrepto parse and aggregate data from local JSONL session logs. These commands are used for their intended purpose of data processing and search. - [DATA_EXPOSURE]: The skill accesses historical conversation transcripts stored in
~/.openclaw/agents/<agentId>/sessions/. While these files contain user and assistant message history, the access is restricted to the local filesystem and is the primary function of the skill to provide historical context. - [PROMPT_INJECTION]: The skill processes historical session data which could contain untrusted input from previous interactions (Indirect Prompt Injection surface). However, the skill only performs text-based searching and structured analysis using
jq, without executing or evaluating the content of the logs, which minimizes risk.
Audit Metadata