Security audit
Security checks for vulnerabilities and agentic risk
The skill is mostly transparent documentation for shared agent memory, but it also documents persistent cross-agent storage and public GitHub publishing with specific account credentials without enough scoping or user-confirmation guidance.
Review this skill before installing if your sessions may contain private, regulated, or secret information. Use it only with content you intend to persist in a shared vault and semantic index, and do not let an agent run the GitHub publishing commands or use the named SSH/gh credentials without explicit review of exactly what will be pushed publicly.
The skill instructs agents to persist valuable ideas, observations, and problems into shared memory and auto-index them, but does not prominently warn that this creates durable storage and semantic retrieval of agent-generated content. Sensitive data, credentials, internal discussions, or regulated information could be captured and become discoverable later across tools and sessions.
The skill states that OpenClaw writes to shared vault/files and Claude Code reads them in later sessions, but it does not clearly warn that anything written there becomes visible across agents and persists beyond the immediate interaction. This can cause cross-agent disclosure of confidential prompts, outputs, or user data through a shared filesystem channel.
The skill’s stated purpose is documenting a shared-memory stack, but it also includes instructions to initialize git repositories and publish skills to GitHub. That expands the skill’s operational scope into external networked actions and code publication, which can lead an agent or user to exfiltrate local content or publish unintended material under an authenticated account.
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
Both agents publish to GitHub under nerua1. SSH key and gh are configured.
# Claude Code skills
cd ~/.claude/skills/vault/SKILL_NAME
git init && git add . && git commit -m "feat: SKILL_NAME v1.0.0"
gh repo create SKILL_NAME --public --source . --remote origin --push
The documentation reveals specific credential and account configuration details, including the SSH key path, authenticated GitHub identity, and git config values. Even without exposing private key material, this materially lowers the barrier for misuse by telling an agent exactly which credentials exist and how to invoke them for outbound actions.
No suspicious patterns detected.