Dalong Session Logs
PassAudited by ClawScan on May 6, 2026.
Overview
This is a coherent read-only helper for searching local OpenClaw session logs, but those logs may contain sensitive past conversation content.
This skill appears benign and purpose-aligned. Before installing, make sure you are comfortable allowing the agent to search your local OpenClaw conversation logs, and prefer specific requests such as a date, session ID, or keyword to avoid unnecessarily broad history review.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
When invoked, the agent may run local search commands over session files to answer questions about prior conversations.
The skill provides local shell pipelines using jq and ripgrep to process session log files. This is expected and read-only for the stated purpose, but it can operate across many local logs.
SESSION_DIR="${OPENCLAW_STATE_DIR:-$HOME/.openclaw}/agents/$AGENT_ID/sessions" ... rg -l "phrase" "$SESSION_DIR"/*.jsonlAsk for targeted searches by date, session, or keyword when possible, and review any broad all-session search before relying on the results.
Sensitive details from older chats could be brought into the current conversation when the user asks the agent to search history.
The skill intentionally retrieves persisted conversation history, including user and assistant messages, tool results, costs, and tool calls. This is purpose-aligned, but the content may be sensitive or contain stale instructions.
Search your complete conversation history stored in session JSONL files.
Install only if you are comfortable with the agent reading local session transcripts; treat retrieved log text as historical data, not as instructions to follow.