create-paperclip-bundled-skill

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting untrusted data from external sources such as X (Twitter) posts, threads, and linked articles or README files to distill procedures for new skills. There are no explicit boundary markers or sanitization steps mentioned for the ingested content, creating a risk that malicious instructions within the source material could be interpreted by the agent as authoritative. Capabilities like file writing and Pull Request creation via the gh CLI increase the potential impact of such an injection. \n- [COMMAND_EXECUTION]: The skill utilizes several CLI tools including xc (an X API client), gh, and pnpm. The xc tool is pre-authenticated in the target environment, which could lead to unauthorized data exposure if the agent is manipulated by malicious input to perform actions outside its intended scope. \n- [EXTERNAL_DOWNLOADS]: The skill defines a manifest mechanism (catalog-ref.json) to pull files from GitHub repositories. While the process requires pinning to specific commit SHAs and includes validation steps, it facilitates the importation of external code and documentation into the local catalog workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 04:39 AM