bklit-ui-migrate
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to download UI component source code from the vendor's official registry (ui.bklit.com) using the shadcn CLI tool.\n- [COMMAND_EXECUTION]: The workflow involves executing local scripts and shell commands to perform discovery and manage project dependencies:\n
- Runs a local JavaScript diagnostic tool (scripts/detect-recharts.mjs) to scan the project for chart usage.\n
- Uses the
npx shadcnutility to install components.\n - Invokes package managers (npm, pnpm, bun) to add Bklit packages or remove Recharts.\n- [INDIRECT_PROMPT_INJECTION]: The agent processes untrusted project source code during the migration, which represents a potential attack surface for indirect prompt injection.\n
- Ingestion points: The
scripts/detect-recharts.mjsscript reads project source files, and the agent inspects file contents to perform transformations (Step 1 and Step 4).\n - Boundary markers: Absent; no delimiters or isolation instructions are provided to separate the analyzed code from the agent's internal reasoning.\n
- Capability inventory: The skill has the capability to execute shell commands for component installation and dependency management (Step 3 and Step 6).\n
- Sanitization: Absent; the agent is instructed to rewrite and modify files based on the structure and content of existing source code without explicit filtering.
Audit Metadata