stock-analysis
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves financial data, news headlines, and social media trends from established services such as Yahoo Finance, CoinGecko, Google News, and Reddit.
- [COMMAND_EXECUTION]: Scripts use
subprocess.runto call thebirdCLI for Twitter sentiment analysis anduvfor executing internal Python modules. These calls are part of the core functionality and utilize hardcoded or validated arguments to prevent shell injection. - [CREDENTIALS_UNSAFE]: The documentation identifies the need for authentication tokens for optional social media integration and instructs the user to store them in a
.envfile, which is a standard and recommended practice for local credential management. - [DATA_EXFILTRATION]: Network operations are restricted to fetching information from well-known domains relevant to the skill's purpose. There is no evidence of sensitive local data being transmitted to unauthorized external servers.
- [REMOTE_CODE_EXECUTION]: No patterns of downloading and executing arbitrary remote scripts were identified. External tools and dependencies are clearly documented and retrieved through standard package registries.
Audit Metadata