Caveat verdict

safe-exec

safe-exec-0-3-2

88
🟢 Trusted
No high-risk patterns surfaced by the deep scan — automated capability review, not behavioral proof.

A command safety layer that intercepts and requires approval for dangerous shell commands; clones from a named GitHub repo with no credential exfiltration, no external callbacks, and the skip functionality is documented as agent-mode convenience with audit logging intact.

⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.

Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

0
security
100
transparency
80
maintenance

Findings (13)

Pattern match critical

Recursive delete from root or home — destructive command

SKILL.md · code · rm -rf /

Pattern match critical

Pipe-to-shell pattern (curl | sh) — supply chain attack vector

docs/BLOG_EN.md · code · curl http://example.com/install.sh | bash

Pattern match critical

Pipe to bash — executes piped content as shell commands

docs/BLOG_EN.md · code · | bash

Pattern match high

Accesses OpenClaw config/secrets directly

docs/FIX_REPORT_v0.1.3.md · code · ~/.openclaw/openclaw.json

Pattern match high

subprocess execution — runs system commands from Python

docs/FIX_REPORT_v0.2.3.md · code · subprocess.run(

Pattern match high

Accesses shell history/config

README_EN.md · code · ~/.zshrc

Pattern match high

Pipe-to-shell pattern (wget | sh)

README-detail.md · prose · downgraded · wget | sh

Pattern match high

Pipe to sh — executes piped content as shell commands

README-detail.md · prose · downgraded · | sh

Pattern match high

Pipe to python — executes piped content as Python code

scripts/safe-exec.sh · prose · downgraded · |python

Pattern match medium

References sudo — requests elevated privileges

docs/BLOG.md · prose · downgraded · sudo

Pattern match medium

Popular HTTP library — network access

docs/BLOG_EN.md · code · got

Pattern match medium

Sets world-executable permissions

README_EN.md · code · chmod 777

Pattern match medium

apt-get install — installs system packages

tools/release.sh · prose · downgraded · apt-get install

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class A/B/D/F). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.

Permissions & capabilities

No declared permissions — minimal attack surface.

Is this flag fair?

Check another skill Browse the registry Auditing your own skills or configs? Use the API