The Agent Skills Directory

[SAFE]: No security threats or malicious patterns were identified across the 125 files analyzed. The skill follows best practices for AI agent extensions and includes several built-in security features:
[AGENT_ORCHESTRATION]: The agent-protocol and chief-of-staff skills implement a robust communication standard between roles, featuring loop prevention (maximum depth limits and circular call blocking) and isolation rules to prevent groupthink during deliberations.
[DATA_PRIVACY]: The skill includes a dedicated anonymization-protocol designed to strip sensitive financial figures and personal names before any theoretical external tool or API invocation.
[LOCAL_PERSISTENCE]: The skill uses a predictable local path (~/.claude/) for persisting company context and decisions, acting as a transparent memory layer for the user's environment.
[CODE_SAFETY]: All 25 included Python analysis tools (e.g., burn_rate_calculator.py, risk_quantifier.py) are implemented using only Python's standard library, minimizing supply chain risks and ensuring no hidden external dependencies are introduced.
[QUALITY_ASSURANCE]: An 'Internal Quality Loop' is enforced, requiring agents to run self-verification, peer-verification, and critic pre-screening (via the executive-mentor role) before presenting strategic recommendations to the user.

c-level-advisor