Token Cost Estimator
PassAudited by ClawScan on May 1, 2026.
Overview
This skill is coherent and purpose-aligned, but its cost estimate works by running a local script that reads all OpenClaw session transcripts.
Before installing or using this skill, be aware that its estimate is based on scanning all local OpenClaw agent session transcripts. The artifacts do not show network sharing, credential access, file modification, or background persistence, but you should narrow the script if you only want selected sessions included.
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.
Running the snippet will execute local code on the machine, but the code shown is limited to reading transcript files and printing aggregate token/cost estimates.
The skill asks the user or agent to execute a local Python script. This is expected for a transcript-based cost estimator and no hidden install-time execution is shown.
Run this Python script to analyze all agents:
Review the script before running it and run it only in an environment where reading local OpenClaw transcripts is acceptable.
The calculation depends on reading past session data across agents, which may include sensitive context from prior conversations.
The script scans all local OpenClaw agent session transcripts. Those transcripts may contain private conversation content or tool outputs, even though the script only prints aggregate counts and costs.
AGENTS_DIR = os.path.expanduser('~/.openclaw/agents') ... for f in glob.glob(os.path.join(sess_dir, '*.jsonl')):Use it only if you are comfortable scanning all local OpenClaw sessions; narrow the path or modify the script if you want to estimate only specific agents or sessions.