ai-screen-vision@1.1.0
CAUTION· install onlyPublisher claims: "AI screen vision and desktop computer control skill for OpenClaw. Let your AI agent see the screen, understand UI elements, and autonomously perform mouse an..." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · 8 of 11 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-19 and scored it 64/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: OS command execution function call detected; OS command execution function call detected; Destructive file system command detected. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 3 scenariosStatic findings
result = subprocess.run(cmd, capture_output=True, text=True, env=env, timeout=10)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.chmod 600 ~/.vnc/passwdFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.chmod +x /usr/local/bin/sv-stopFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.ret = os.system(cmd)Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.subprocess.run([sys.executable, "-m", "pip", "install", "--quiet",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.chmod +x ~/.vnc/xstartupFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.chmod +x /usr/local/bin/sv-startFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.vncserver -kill :1 2>/dev/nullFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.os.system(f"cp {ss_path} {log_ss} 2>/dev/null")Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.subprocess.run(['DISPLAY', '$display', 'scrot', '-z', '-overwrite', '$OUTPUT'], env={'DISPLAY': '$display', 'PATH': '/usr/bin:/bin'})Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, capture_output=True, text=True, env=env, timeout=30)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.kill $(cat /tmp/sv-novnc.pid 2>/dev/null) 2>/dev/nullFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.rm -f /tmp/sv-novnc.pidFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
16 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- OS command execution function call detected≈ +9 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Unclaimed publisher≈ +8 on the totalClaim guitu917: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Touches files outside the project≈ +7 on the totalKeep reads and writes inside the working directory, or declare the exact paths in `filesystem:` and explain why in the README.
- Destructive file system command detected≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.