skills/johnie/skills/stricli/Gen Agent Trust Hub

stricli

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation references @stricli/core and @stricli/auto-complete from the official NPM registry. These packages are maintained by Bloomberg, a well-known and reputable technology organization.
  • [COMMAND_EXECUTION]: Example 8 in the documentation demonstrates using Bun's shell execution API ($) to wrap existing CLI tools like git. This is a standard documentation pattern for CLI wrappers and utilizes Bun's built-in security features for shell argument escaping.
  • [PROMPT_INJECTION]: The framework describes building applications that ingest external input via CLI flags and positional arguments, creating a surface for Indirect Prompt Injection. Ingestion points: CLI parameters defined via buildCommand in SKILL.md and references/examples.md. Boundary markers: The framework uses type-safe parsers and enums but does not implement LLM-specific boundary markers. Capability inventory: Documentation examples across several reference files demonstrate capabilities for file system access and shell command execution. Sanitization: Input is validated and transformed via the framework's parsing layer before being processed by application logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 12:23 PM