moai-lang-php

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill configuration permits the agent to read local files and fetch data from the web, which creates a significant surface for indirect prompt injection when combined with the allowed terminal (Bash) tool.
  • Ingestion points: Read (read_file), WebFetch, and WebSearch tools are explicitly permitted.
  • Boundary markers: The skill does not provide instructions to the agent to treat external data as untrusted or to use delimiters when processing file or web content.
  • Capability inventory: The agent is authorized to use a terminal (Bash) for command execution, posing a risk if instructions are injected into read data.
  • Sanitization: There is no evidence of validation or sanitization logic to filter instructions from data retrieved via WebFetch or file reads.
  • [NO_CODE]: No executable script files (e.g., .py, .js, .sh) are provided in the skill package. The current analysis is based on the permissions and documentation provided in the markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 05:14 PM