session-logs

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs reading and outputting full prior message text from session JSONL files (with jq/rg examples), which can include API keys, tokens, or passwords and therefore requires handling/outputting secrets verbatim.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s “Search your complete conversation history stored in session JSONL files” reads local session transcript text from ~/.openclaw/agents/<agentId>/sessions/ (including outsider user-authored messages), which can be ingested when the workflow searches those logs for historical context.