Skillv1.0.0

ClawScan security

Agent Comm Monitor · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 22, 2026, 4:58 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requests and runtime instructions are internally consistent with a monitoring/reporting tool for inter-agent communication, but it will read session histories and write logs to workspace files so you should consider privacy and logging policy before enabling it.
Guidance: This skill appears to do what it says: send test messages, check session history, and produce a report. Before installing, confirm you are comfortable with the agent having access to sessions_send/sessions_history and with the skill writing log entries to MEMORY.md and HEARTBEAT.md (these may contain message contents). Consider: run it with explicit invocation only if you want to avoid automatic monitoring; verify where those log files are stored and how long they are retained; redact or avoid logging sensitive message content; and test in a non-production environment first. If you need stronger guarantees, ask the skill author how logs are protected and whether logs are shared externally.

Purpose & Capability: okName/description match the instructions: the SKILL.md calls the platform session APIs (sessions_send, sessions_history) and targets specific agent session keys to test and report cross-agent messaging — this is proportional to a communication-monitoring skill.
Instruction Scope: noteInstructions stay on-topic (send test messages, check session history, interpret replies). They also instruct the agent to write anomalies to MEMORY.md and HEARTBEAT.md; reading session histories and persisting logs is expected for monitoring but has privacy implications because it accesses message contents and creates persistent records.
Install Mechanism: okNo install spec and no code files — instruction-only behavior is low-risk from an installation perspective (nothing is downloaded or written to disk by an installer).
Credentials: okThe skill requests no environment variables or external credentials. Its operational needs (access to sessions_* APIs and the agent workspace) are appropriate for its stated purpose.
Persistence & Privilege: notealways is false and the skill is not force-included. However, it explicitly directs writing to MEMORY.md and HEARTBEAT.md (persistent workspace files) and recommends logging after repeated failures — this grants it the ability to create/modify local logs which the user should be aware of.