contribute-catalog

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx to download and run packages including hyperframes, oxfmt, and tsx. Additionally, the provided templates load external JavaScript libraries (gsap, three.js) from cdn.jsdelivr.net and typography from fonts.googleapis.com.
  • [COMMAND_EXECUTION]: Multiple shell commands are invoked throughout the workflow, including git for branch and commit management, gh for pull request operations, and the hyperframes tool for linting. The skill also executes local repository scripts such as scripts/upload-docs-images.sh and scripts/generate-catalog-pages.ts.
  • [DATA_EXFILTRATION]: The skill is designed to move local data to external endpoints as part of the submission process, specifically publishing to hyperframes.dev and pushing source code to GitHub repositories.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface. Ingestion points: User-provided descriptions and visual reference data are collected in the 'Clarify' step of SKILL.md to define the parameters of the new registry item. Boundary markers: Absent; the instructions do not implement delimiters or 'ignore' directives for user-provided strings. Capability inventory: The skill is capable of writing files to the local directory and executing shell commands (via npx, git, and gh). Sanitization: No input validation or escaping mechanisms are described for the user-supplied text before it is used in file generation or command metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 02:54 PM
Security Audit — agent-trust-hub — contribute-catalog