Skills
Navigate
skills/athola/claude-night-market/mcp-code-execution/Gen Agent Trust Hub

mcp-code-execution

Warn

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The file modules/mcp-patterns.md instructs the agent to run a Python script at tools/extracted_tool.py. This file is not included in the provided skill bundle, making its execution behavior unverifiable and dependent on the presence of unverified local files.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by processing large, potentially untrusted datasets and workflows.
  • Ingestion points: Data ingestion occurs through the input_data parameter in modules/mcp-coordination.md and workflow descriptions in SKILL.md.
  • Boundary markers: The skill lacks documentation of delimiters or instructions to ignore commands within the data.
  • Capability inventory: The skill has the ability to execute code and coordinate multiple subagents.
  • Sanitization: No data validation or sanitization routines are present for processed inputs.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 29, 2026, 04:49 PM