skills/jeremylongshore/claude-code-plugins-plus-skills/tracking-crypto-portfolio/Gen Agent Trust Hub
tracking-crypto-portfolio
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE_&_EXFILTRATION]: The skill performs network requests to the CoinGecko API (
api.coingecko.com) to fetch real-time cryptocurrency prices. This is a well-known service and the operations are restricted to fetching public market data. No sensitive local data or credentials are transmitted. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided portfolio JSON files.
- Ingestion points:
scripts/portfolio_loader.pyreads holdings from local JSON files. - Boundary markers: The skill treats the input as structured JSON and enforces a strict schema for keys like
coin,quantity, andcost_basis. - Capability inventory: The skill uses
requestsfor network access (CoinGecko) and writes output to local files (JSON/CSV exports). - Sanitization:
PortfolioLoadervalidates and casts input values (e.g., converting quantity to floats) before they are used in calculations or formatted for output, effectively neutralizing injection risks into the processing logic. - [EXTERNAL_DOWNLOADS]: The skill requires the
requestsPython package, which is a standard, well-known library for HTTP operations. It does not download or execute any unknown external scripts. - [COMMAND_EXECUTION]: The skill uses scoped Bash tool access (
crypto:portfolio-*) to execute its own Python scripts. This implementation follows the principle of least privilege by restricting shell execution to a specific namespace.
Audit Metadata