c-level-advisor
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified in the skill instructions or the provided analysis scripts.
- [EXTERNAL_DOWNLOADS]: The skill is installed via a standard
npxcommand from the author's repository. This installation method is a standard practice for the intended platform. - [COMMAND_EXECUTION]: The collection includes 25 local Python scripts for data analysis (e.g.,
burn_rate_calculator.py,tech_debt_analyzer.py). These scripts use only the Python standard library and do not perform any hidden network or system-level operations. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it is designed to ingest and analyze company documents such as strategy plans and OKR files.
- Ingestion points: advisor scripts and the
chief-of-staffrouting logic process external files (e.g.,strategy-doc.md) within the project directory. - Boundary markers: The skill implements an 'Internal Quality Loop' and 'User Communication Standard' in
agent-protocol/SKILL.mdto maintain control over the agent's behavior and prevent instructions in user data from overriding system goals. - Capability inventory: The agent can execute included scripts and read/write local files within the project and the
~/.claude/context directory. - Sanitization: A formal
anonymization-protocol.mdis provided to strip sensitive business metrics, customer names, and employee identities before any external tool calls or web searches are conducted.
Audit Metadata