skills/fradser/dotclaude/image/Gen Agent Trust Hub

image

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides the agent with specific shell commands for image processing and inspecting web assets. These include utilities for local file manipulation such as cwebp, mogrify, and jpegoptim, as well as curl combined with grep to scan external URLs for image assets.\n- [PROMPT_INJECTION]: The skill is designed to ingest data from external project files to inform its behavior, which creates a potential surface for indirect prompt injection.\n
  • Ingestion points: The skill instructs the agent to read marketing context files including .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md if present.\n
  • Boundary markers: There are no instructions provided to treat the content of these files as untrusted data or to ignore instructions contained within them.\n
  • Capability inventory: The skill leverages capabilities for shell command execution (cwebp, mogrify, jpegoptim, curl, screencapture) and filesystem access to process data.\n
  • Sanitization: No sanitization, validation, or filtering of the content within the marketing context files is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 12:30 AM
Security Audit — agent-trust-hub — image