Exec Inspector

WarnAudited by ClawScan on May 10, 2026.

Overview

Exec Inspector is mostly aligned with viewing OpenClaw exec history, but it asks the agent to run helper scripts that are not included for review and supports a persistent background monitor of all exec commands.

Review this skill carefully before installing. Its core idea is reasonable, but do not let it run the referenced ~/.openclaw/scripts helpers or start daemon mode unless you have inspected those scripts and are comfortable with continuous monitoring of exec command history.

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.

What this means

The agent could run whatever script currently exists at that local path, not necessarily reviewed code from this skill.

Why it was flagged

The skill directs the agent to execute local helper scripts, but the supplied review context says this is an instruction-only skill with no install spec and no code files. The behavior of those scripts cannot be verified from the artifacts.

Skill content
AI 应该立即执行相应命令 ... `~/.openclaw/scripts/exec-history.sh list`
Recommendation

Only install or use this after verifying the contents and origin of ~/.openclaw/scripts/exec-history.sh and ~/.openclaw/scripts/exec-monitor-daemon.sh, or require the skill to include reviewed helper code and installation steps.

What this means

Future commands may be continuously captured and written to logs, potentially including sensitive command arguments or operational details.

Why it was flagged

The skill documents a background daemon that continues monitoring future exec tool use after it is started.

Skill content
后台守护进程 - 自动捕获所有 exec 执行并实时输出 ... 从此刻起,OpenClaw 每次执行 exec 命令时,都会自动在日志中实时输出!
Recommendation

Use daemon mode only when intentionally monitoring, confirm where logs are stored, stop it after use, and prefer an implementation that clearly documents retention, scope, and cleanup.

What this means

Command history or session details may be displayed to the model/user or saved to a file.

Why it was flagged

The skill reads persistent OpenClaw session logs and can export history. That is expected for an exec-history inspector, but these logs may contain sensitive commands, arguments, and tool metadata.

Skill content
Session 文件位置 ... `~/.openclaw/agents/main/sessions/` ... `exec-history export`
Recommendation

Review outputs before sharing them, avoid exporting sensitive histories unnecessarily, and consider limiting inspection to a specific session when possible.