Security audit
Security checks for vulnerabilities and agentic risk
This skill provides disclosed project decision memory behavior with user-gated validation and no evidence of hidden execution, exfiltration, or destructive actions.
Before installing, understand that this skill is meant to remember project decisions across sessions. Use it for repositories or long-running projects where that is helpful, and only validate memories when you are comfortable making them durable project context.
The instruction to use the skill whenever the user is working on a repo, software project, agent workflow, plugin, MCP server, or multi-step implementation is very expansive and lacks clear exclusion conditions. This makes activation scope ambiguous and likely to overlap with a large share of normal engineering conversations.
Conditions such as 'the task is part of an ongoing project' and 'the user asks to continue, refine, implement, or fix something' are common in everyday development requests and do not clearly distinguish when retrieval should not occur. The list also lacks negative examples or tighter constraints, so the activation boundary is unclear.
No suspicious patterns detected.
