Session Cost
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
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.
Session logs can reveal historical agent activity, model usage, costs, timestamps, and session identifiers, even if the visible code focuses on metrics rather than message content.
The skill intentionally reads persistent OpenClaw session logs across agents to calculate usage and cost metrics.
By default, scans all agents in `~/.openclaw/agents/`. Each agent's sessions are read from `~/.openclaw/agents/<name>/sessions/`.
Use the provided filters such as --agent and --offset when you want a narrower report, and avoid sharing detailed or JSON output if usage history or session identifiers are sensitive.
If a broad custom directory is supplied, the skill may process more local JSONL files than intended.
The script accepts a user-supplied path and recursively reads matching .jsonl files under that directory.
args.path = argv[++i]; ... if (entry.isDirectory()) { walk(fullPath); } else if (entry.isFile() && entry.name.endsWith('.jsonl')Only pass trusted, narrowly scoped directories to --path; prefer the default OpenClaw sessions location or explicit --agent/--offset filters for routine cost reports.