deserialization-php

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download the phpggc tool from its official GitHub repository. This is a standard and expected action for the described penetration testing use case.
  • [COMMAND_EXECUTION]: The instructions utilize common CLI tools such as php, curl, and python3 to generate payloads and interact with targets. These actions are within the scope of the skill's stated purpose for security testing.
  • [PROMPT_INJECTION]: The skill processes data from a state MCP server via get_state_summary(), which creates a surface for indirect prompt injection from potentially untrusted engagement logs.
  • Ingestion points: MCP server engagement state (SKILL.md)
  • Boundary markers: Absent
  • Capability inventory: Shell command execution and network requests via php, python3, and curl (SKILL.md)
  • Sanitization: None identified for processing external state data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 09:15 AM
Security Audit — agent-trust-hub — deserialization-php