mupengism

PassAudited by ClawScan on May 1, 2026.

Overview

This is a documented, instruction-only memory framework with no code or credentials, but it intentionally creates persistent workspace memory that can affect future sessions.

Install this only if you want the agent to maintain persistent workspace memory. Review SOUL.md, MEMORY.md, and memory/*.md like long-term notes that can shape future behavior, keep them scoped to trusted workspaces, and do not store secrets or sensitive personal data there.

Findings (2)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Future agent sessions may follow instructions or assumptions stored in the workspace memory files.

Why it was flagged

The skill intentionally makes a persistent file part of the agent's future session context, so stale, incorrect, or manipulated content in that file could influence later behavior.

Skill content
Configure AGENTS.md to load SOUL.md every session
Recommendation

Keep SOUL.md and memory files in a trusted workspace, review changes periodically, and avoid letting untrusted content be written into persistent memory.

What this means

If users store private details in these files, they may be reloaded into future sessions or exposed in shared contexts.

Why it was flagged

The memory system may contain personal values and context, and the artifact itself acknowledges that secrets and sensitive personal information should not be recorded.

Skill content
MEMORY.md는 민감하다 ... 형님의 가치관, 개인 맥락 포함 ... 비밀번호, API 키, 니모닉 ... 민감한 개인정보
Recommendation

Do not store passwords, API keys, wallet phrases, or sensitive personal data in the memory files; avoid loading them in group chats or external sessions.