make-font
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The script
scripts/find_foundation.pyfetches font metadata and open-license font files from well-known technology services, specifically Fontsource (api.fontsource.org) and the official Google Fonts repository on GitHub (api.github.com,raw.githubusercontent.com). These downloads are used solely to provide valid open-source font foundations for the typeface generation process. - [COMMAND_EXECUTION]: The skill utilizes shell command execution for necessary administrative and build tasks:
scripts/find_foundation.pyinvokescurlviasubprocess.runas a fallback to download assets when standard Python SSL/TLS libraries fail.scripts/check_font.pyruns thefontbakerycommand-line tool to perform industry-standard quality checks on generated fonts.scripts/render_book.mjsexecutesnpmto determine the system's package root and launches a Chromium instance via Playwright to render the final font book PDF.- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data which presents a surface for indirect prompt injection attacks.
- Ingestion points: The scripts
measure_font.pyandmeasure_image.pyingest user-supplied font files (.ttf,.otf,.woff2) and screenshots (.png). - Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present in the scripts when processing font metadata or performing OCR on images.
- Capability inventory: The skill possesses the ability to write to the file system, perform network requests to trusted domains, and execute subprocesses.
- Sanitization: Although the ingested data is used to calculate numeric font properties, the skill does not perform sanitization of string metadata or screenshot text to prevent potential adversarial instructions from being interpreted by the agent.
Audit Metadata