stack-modernization
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages shell commands to perform its core modernization tasks, specifically using the Bun package manager and Git.
- Evidence includes the use of
bun add,bun remove,bun outdated, andbun run testto manage dependencies and verify project integrity. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect injection through external data sources like web search results and package manager outputs.
- Ingestion points: The skill uses
WebSearchto retrieve current package versions and migration guides from external websites (npm, GitHub). - Boundary markers: The instructions mandate a 'confirmation gate' where a modernization plan must be approved by the user before execution.
- Capability inventory: The skill has authorized access to execute
bunandgitcommands, allowing it to modify files and install packages. - Sanitization: The skill follows a strict incremental process, running tests after every single package change to ensure that any regressions are immediately identified and can be rolled back.
Audit Metadata