script-portability-acrazie
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill recommends using
node -e(inline execution) and Bun shell commands as portable alternatives to common polyfills likerimraf. This involves generating and executing code strings at runtime based on provided templates.- [COMMAND_EXECUTION]: The skill workflow includes verification steps where the agent executes tests on the refactored scripts. It includes significant guardrails, such as requiring explicit user permission for all executions and discouraging the use of unverified or unfamiliar scripts.- [INDIRECT_PROMPT_INJECTION]: The skill audits third-party scripts and manifests, which presents a surface for indirect prompt injection where malicious instructions in the audited files could attempt to influence agent behavior. - Ingestion points: Manifests like
package.jsonand various shell scripts are read during Phase 1 (Inventory). - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' markers when reading these files.
- Capability inventory: The skill can modify project files, synchronize lockfiles, and execute shell commands for verification.
- Sanitization: The skill explicitly recommends sanitizing output (e.g., using parameterized APIs and template escaping) to prevent downstream injection, though it does not specify sanitization for the input audit data.
Audit Metadata