stitch-svelte-components

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses a bash script to download content from dynamic URLs provided by external tools.\n
  • Evidence: The script scripts/fetch-stitch.sh executes curl -L to download design assets from a provided URL to a local file path.\n- [COMMAND_EXECUTION]: The skill requires the execution of shell scripts and package manager commands to perform its tasks.\n
  • Evidence:\n
  • Execution of scripts/fetch-stitch.sh via bash to retrieve design data.\n
  • Execution of npm install and npm run dev for environment setup and verification.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes HTML data from external URLs to generate code.\n
  • Ingestion points: The skill downloads HTML from [htmlCode.downloadUrl] (Step 1.3).\n
  • Boundary markers: None identified; the skill does not instruct the agent to ignore instructions that might be embedded within the downloaded HTML content.\n
  • Capability inventory: The skill has access to Bash, Read, and Write tools, enabling it to execute commands and modify the filesystem based on interpreted content.\n
  • Sanitization: No explicit sanitization or validation of the downloaded HTML content is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 06:44 PM
Security Audit — agent-trust-hub — stitch-svelte-components