The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill contains a 'Self-Evolution Subsystem' (defined in references/self-evolution.md) that allows the agent to autonomously update its own SKILL.md and reference files based on task performance and feedback. It utilizes file-writing tools to perform these updates. Although governed by safety levels and human approval requirements for core changes, this autonomous instruction-modification pattern warrants monitoring.
[PROMPT_INJECTION]: The skill's primary function is to generate new agent skills based on untrusted user input, creating a surface for indirect prompt injection. The skill mitigates this through a structured validation phase and safety-oriented design principles (references/agent-evaluation-guardrails.md), but the ingestion of arbitrary requirements remains a potential vector.

architect