skills/arthjean/skills/coss-particles/Gen Agent Trust Hub

coss-particles

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides a directory of over 480 URLs pointing to UI component manifests hosted at https://coss.com/ui/r/. These links are used to retrieve the source code for copy-paste-ready UI patterns.
  • [COMMAND_EXECUTION]: The documentation provides a command (node apps/ui/scripts/generate-particle-index.cjs) intended for users or agents to update the local index. This is a standard development workflow pattern within the context of the supported project repository.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to fetch and process source code from external JSON manifests. If the content on the remote server were compromised, it could influence the agent's behavior.
  • Ingestion points: Remote JSON manifests at https://coss.com/ui/r/<particle-name>.json as specified in SKILL.md.
  • Boundary markers: None present; the agent is instructed to fetch and adapt the code directly.
  • Capability inventory: No specific tool capabilities or advanced script execution are defined in this markdown-only index.
  • Sanitization: No sanitization or validation logic is provided for the fetched remote content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 08:59 PM