The Agent Skills Directory

[Unverifiable Dependencies & Remote Code Execution] (CRITICAL): File references/mcp-setup.md instructs users to install Homebrew using curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh | bash. Piping a remote script directly to a shell from an untrusted source (Homebrew is not in the TRUST-SCOPE-RULE list) is a critical security vulnerability that can lead to system compromise.
[Unverifiable Dependencies & Remote Code Execution] (HIGH): The skill directs users to install and run code using npx skills add and npx -y octagon-mcp. These commands download and execute code from an unverified third-party ('OctagonAI') which is not a trusted source.
[Indirect Prompt Injection] (LOW): The skill ingests data from external SEC filings. 1. Ingestion points: SEC 10-K and 10-Q filings are retrieved via the octagon-agent tool. 2. Boundary markers: None identified. 3. Capability inventory: The octagon-agent provides analysis and research summaries. 4. Sanitization: No evidence of sanitization of filing content.
[Data Exposure & Exfiltration] (LOW): The configuration instructions recommend setting OCTAGON_API_KEY in environment variables or configuration files. While standard for many tools, this exposes the credential to any process able to read the environment or local configuration.

sec-cash-flow-review