ClawAudit verdict

eva-soul

eva-soul-by-openclaw

88
🟢 Trusted
Low risk — reviewed by ClawAudit, behavior matches stated purpose

This skill provides a local AI personality/emotion/memory system for OpenClaw with all data stored in local JSON files; it makes no external network calls, accesses no secrets, and behavior matches its stated purpose as a cognitive plugin.

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

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

55
security
100
transparency
100
maintenance

Findings (4)

Pattern match critical

Recursive delete from root or home — destructive command

SKILL.md · code · rm -rf ~

Pattern match medium

Uses exec() — may execute shell commands

lib/cognition/cognition.js · prose · downgraded · exec(

Pattern match low

Accesses sensitive environment variables

hooks/postResponse.js · prose · downgraded · process.env.SILICONFLOW_API_KEY

Pattern match low

Popular HTTP library — network access

verify.js · prose · downgraded · got

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