Security audit
Security checks for vulnerabilities and agentic risk
The skill is a coherent memory and context-loading guide, but it tells agents to read broad personal/workspace context and persist new session information by default without clear user approval or retention controls.
Install only if you are comfortable with the agent proactively reading agent-readable workspace memory, identity, task, tool, and skill files, and writing session notes for future use. Keep secrets out of those files, review memory/task updates regularly, and prefer explicit approval before storing personal, customer, security, or business-sensitive information.
SKILL.md:39Excessive Proactive Collection of Local Context
references/session-checklist.md:56Automatic Persistent Retention of Session Information
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
**Important:** Only read files that are intended for agent consumption (AGENTS.md, TOOLS.md, SOUL.md,
USER.md, memory/, skills/, tasks/). Do NOT proactively read `.env`, `.secrets/`, credentials files,
or other sensitive locations. Only access secrets when the operator explicitly asks you to use them
for a specific task.
### Why this matters:
The checklist instructs the agent to proactively read multiple workspace identity and memory files before the user has provided context or consent, creating a broad default data-access pattern. This can expose sensitive or irrelevant information from prior sessions, tasks, or user metadata and violates least-privilege principles by normalizing unnecessary prefetch of local context.
No suspicious patterns detected.