Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
my_soul_upgrade_skill
v1.0.1Manage and synchronize global and agent-specific Soul definitions using a two-layer template system with automated rebuild scripts.
⭐ 0· 94·1 current·1 all-time
by@canonxu
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The described capability (synchronizing a global SOUL and per-agent templates) aligns with the files and paths the SKILL.md references. However, it hardcodes /home/admin paths and relies on an unspecified build script in the user's home; this is plausible for the stated purpose but the absolute admin path and missing script are noteworthy.
Instruction Scope
SKILL.md directs editing of files under /home/admin/.openclaw and running python3 ~/.openclaw/scripts/build_all_souls.py. The build_all_souls.py script is not included in the skill bundle, so the instructions require execution of an external, unknown script that could read/modify any agent workspace — broader than what the skill package provides and potentially dangerous.
Install Mechanism
This is an instruction-only skill with no install spec and no bundled code to execute, which minimizes installation risk. The risk arises from the runtime instruction to execute a script that is not part of the package.
Credentials
No environment variables or credentials are requested (appropriate). However, the SKILL.md assumes filesystem access to /home/admin/.openclaw and agent workspaces without declaring or justifying privileged access to an admin home directory.
Persistence & Privilege
The skill does not request permanent/always-present privileges and does not modify other skills. Still, its instructions allow invoking a script that rebuilds SOUL files for all agents, which is a high-impact operation if executed by the agent autonomously.
What to consider before installing
Do not run the recommended build_all_souls.py until you inspect its contents and confirm its provenance. The SKILL.md tells the agent to run a Python script located in ~/.openclaw/scripts that is not included in the skill bundle — that script could perform arbitrary operations across all agent workspaces. Before installing or invoking the skill: (1) ask the publisher to provide the build_all_souls.py source or a trusted release link; (2) review the script for network calls, filesystem modifications, and privilege escalation; (3) prefer running it manually in a sandbox or as a non-admin user and back up SOUL files; (4) verify whether /home/admin is the correct path for your deployment or whether the instructions should use the agent's workspace variables; (5) if you cannot inspect the script or confirm the source, avoid granting the agent permission to run it autonomously.Like a lobster shell, security has layers — review code before you run it.
latestvk97anbm2tg7k0mpt9rvqd6pdkd83x37g
License
MIT-0
Free to use, modify, and redistribute. No attribution required.