Multi-viewpoint Debates
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill mostly matches its stated debate purpose, but its helper script prints copy-paste shell commands with the user's topic inserted unsafely, and it stores potentially private debate history.
This skill appears generally purpose-aligned for running decision debates, but review the generated shell commands before running them. Avoid untrusted or shell-like characters in the debate topic, and be careful about what personal or business information you save in the archive.
Findings (3)
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 a debate can create several separate model sessions and associated outputs.
The skill's main workflow launches multiple sub-agent sessions. This is central to the skill and disclosed, but users should understand it creates additional agent activity and transcripts.
clawdbot sessions_spawn --task "You are Elon Musk ..."
Use the skill only when you intentionally want multiple perspectives, and review the generated session outputs before archiving or sharing them.
A malformed or untrusted debate topic could cause a copied command to do more than spawn a debate session, including running local shell commands.
The script prints shell commands for the user to copy and run, placing the raw topic inside a double-quoted --task command. A topic containing quotes, command substitutions like $(...), backticks, or newlines could alter the copied command or execute unintended shell code.
TOPIC="${1:-Should I continue with this project?}" ... echo "Copy and run these three commands in sequence:" ... echo "$TOPIC"Do not paste generated commands for topics containing shell metacharacters. The skill author should build the task as a safely quoted argument, pass it through stdin or a temporary file, or directly invoke clawdbot using an argument array instead of printing raw shell commands.
Sensitive decisions, project metrics, URLs, and reasoning could remain in a searchable local archive and be reused later.
The skill encourages copying agent transcript content into a persistent archive. That is aligned with its decision-history purpose, but the stored debates may contain private business or personal context.
Copy full responses from Clawdbot output or fetch from session transcripts: ~/.clawdbot/agents/main/sessions/[session-id].jsonl ... Save to Archive ... Full responses
Avoid putting secrets in debate prompts, store archives in a private location, and periodically review or delete old debate files you no longer need.