session-logs

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Employs shell pipelines with jq, rg, and awk to process session data.
  • [DATA_EXFILTRATION]: Accesses personal conversation history stored in ~/.openclaw/agents/<agentId>/sessions/. While this path is specific to the agent framework, it contains sensitive user data.
  • [PROMPT_INJECTION]: Processing historical data creates an attack surface for indirect prompt injection.
  • Ingestion points: Session logs in ~/.openclaw/agents/<agentId>/sessions/*.jsonl.
  • Boundary markers: Absent; no delimiters are used when extracting text from logs.
  • Capability inventory: Shell command execution via jq and rg for data processing.
  • Sanitization: None; historical content is read directly without filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 12:13 AM