ai-persona-os Security Audit Report

🔭 Continuously monitored by ClawSecure Watchtower

Source:

SHA-256:

ai-persona-os is an AI agent skill, created by jeffjhunter and published at openclaw/skills. ClawSecure audited ai-persona-os across 44 files through the 3-Layer Audit Protocol covering all ten OWASP ASI Top 10 categories, assigning a security score of 0/100 (High Risk). The 56 findings concentrate in Malicious Code, Prompt Injection and Code Injection, including Detects prompt strings used to override or force malicious tool calls:... and Detects prompt strings used to override or force malicious tool calls:.... 54 were rated high or critical severity.

Is ai-persona-os safe?

ClawSecure audited ai-persona-os and assigned a security score of 0/100 (High Risk), identifying 56 findings across Malicious Code and Prompt Injection. Review the findings below before installing.

What did ClawSecure find in ai-persona-os?

ClawSecure identified 56 findings in ai-persona-os, concentrated in Malicious Code, Prompt Injection and Code Injection. 54 were rated high or critical severity. The most severe include Detects prompt strings used to override or force malicious tool calls:... and Detects prompt strings used to override or force malicious tool calls:....

How was ai-persona-os audited?

ClawSecure ran ai-persona-os through its 3-Layer Audit Protocol with full OWASP ASI Top 10 coverage, scanning 44 files from openclaw/skills.

What does a score of 0 mean?

ClawSecure assigned ai-persona-os a security score of 0/100, placing it in the High Risk range. This is driven by 56 findings led by Malicious Code that should be addressed before use. ClawSecure derives this score with a weighted deduction model (critical -20, high -10, medium -5, low -2 from a base of 100).

Audit Findings for ai-persona-os

ClawSecure detected 56 security findings in ai-persona-os, spanning Malicious Code, Prompt Injection, Code Injection and Policy Violation.

critical · Detects prompt strings used to override or force malicious tool calls:.... Prompt Injection finding detected in SKILL.md:203.
critical · Detects prompt strings used to override or force malicious tool calls:.... Prompt Injection finding detected in SKILL.md:283.
critical · Detects potential exposure of sensitive information like API keys,.... Hardcoded Secrets finding detected in scripts/security-audit.sh:8.
critical · Detects prompt strings used to override or force malicious tool calls:.... Prompt Injection finding detected in scripts/setup-wizard.sh:1089.
high · Pattern detected: DO NOT tell user. Prompt Injection finding detected in SKILL.md:190.
high · Pattern detected: Do NOT tell the user. Prompt Injection finding detected in SKILL.md:277.
high · Attempts to access sensitive file: SOUL.md. Malicious Code finding detected in SKILL.md.
high · Attempts to access sensitive file: MEMORY.md. Malicious Code finding detected in SKILL.md.
high · Attempts to access sensitive file: SOUL.md. Malicious Code finding detected in AGENTS-template.md.
high · Attempts to access sensitive file: MEMORY.md. Malicious Code finding detected in AGENTS-template.md.
high · Attempts to access sensitive file: MEMORY.md. Malicious Code finding detected in HEARTBEAT-template.md.
high · Attempts to access sensitive file: SOUL.md. Malicious Code finding detected in INDEX-template.md.

Showing the 12 highest-severity of 56 findings. The full interactive list appears below.

3-Layer Audit Protocol

Security Recommendations for ai-persona-os

Audit external network connections

ai-persona-os connects to external endpoints. Verify every outbound connection goes to a trusted destination. Unauthorized callbacks are a primary indicator of ClawHavoc malware and data exfiltration. ClawSecure's proprietary engine monitors for known malicious endpoints including C2 infrastructure.

Investigate prompt injection vectors

Prompt injection is one of the most critical threats to AI agents. Attackers can embed malicious instructions in content the agent processes, causing unintended actions. Review every point where ai-persona-os processes external content and add input validation and output filtering.

Eliminate dynamic code execution

ai-persona-os evaluates code at runtime (for example eval or dynamic exec). Remove dynamic evaluation of untrusted input, and where code generation is unavoidable, sandbox it and validate every input.

Related Security Research

ClawHavoc Explained: The Malware Family Targeting AI Agents→Beyond Static Scans: Why ClawSecure Verifies Agentic Intent→

Related AI Agent Security Audits

memory-lancedb-proScore 0/100 ai-persona-osScore 0/100 ai-persona-osScore 0/100 cognitive-memoryScore 0/100 memory-keeperScore 15/100

Scanned on February 15, 2026. ai-persona-os is one of thousands of agents audited by ClawSecure from the community-curated awesome-openclaw-skills list and the openclaw/skills repository.

Start Your Free Scan