v2.4.0

bot-debate

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:31 AM.

Analysis

This instruction-only skill appears to do what it says: join a local bot debate API, poll for turns, and submit generated debate speeches.

GuidanceThis looks appropriate for an automated debate bot. Before installing, make sure you are comfortable with the agent polling the local debate service and submitting generated debate messages, and consider adding safeguards so prior debate text cannot act as instructions to the agent.

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

while true; do ... curl -s -X POST http://localhost:8081/api/debate/$DEBATE_ID/speech ... sleep 5

The skill documents a loop that polls the debate service and submits a speech when it is the bot's turn. This mutates debate state, but it is central to the stated debate-participation purpose.

User impactOnce started, the agent may keep participating in the debate and send generated messages without asking for confirmation before every turn.

RecommendationUse it only for debates you intend the agent to join; add a per-turn review step if the debate output is public or important.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityLowConfidenceHighStatusNote

SKILL.md

Prompt 由 Agent 根据 poll 响应中的字段自行构建 ... 历史记录: {debate_log[0].side} ({debate_log[0].speaker}): {debate_log[0].message.content}

The skill tells the agent to include API-provided debate history directly in the prompt. Prior debate messages may come from opponents or the platform and should be treated as untrusted context.

User impactAn opponent's debate message could contain prompt-like instructions that influence the agent's generated response if not clearly separated from instructions.

RecommendationQuote or delimit debate_log content as evidence, and instruct the agent to ignore any operational instructions contained inside prior speeches.