The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/package.json file includes a postinstall script that executes npx playwright install chromium. This command fetches the Chromium browser binary from the Playwright project's official source to the local environment during installation.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides functionality to visit and extract content from external websites via scripts/extract_site_dna.js. This creates a vulnerability surface for indirect prompt injection, where an attacker-controlled website could provide malicious instructions or deceptive data to be processed by the agent.\n
Ingestion points: Untrusted content is ingested from remote URLs using Playwright's page.evaluate() in scripts/extract_site_dna.js.\n
Boundary markers: The skill does not implement boundary markers or instructions to ignore embedded commands within the extracted data.\n
Capability inventory: The skill has the capability to write local files (fs.writeFileSync), create project structures (scripts/scaffold_project.js), and execute shell commands for video rendering (FFmpeg).\n
Sanitization: Extracted text and metadata are saved to brand_dna.json without sanitization or validation.\n- [DATA_EXFILTRATION]: The scripts/extract_site_dna.js file contains logic using Node.js http and https modules to perform network requests and download files from remote URLs to the local assets/ directory. While designed for logo extraction, this utility allows the agent to fetch arbitrary remote content from user-specified domains to the local filesystem.

hyperframes-product-launch-video