skills/falkicon/mechanic/s-clean/Gen Agent Trust Hub

s-clean

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill processes untrusted external data (WoW addon source code and documentation) to identify dead code and stale references.
  • Ingestion Points: External addon files are analyzed via MCP tools addon.deadcode and docs.stale.
  • Boundary Markers: Absent. The instructions do not specify delimiters for separating the source code from the agent's instructions.
  • Capability Inventory: The skill uses MCP tools for analysis but does not explicitly define file-write or network capabilities; however, it influences the agent's decision-making regarding code deletion.
  • Sanitization: Absent. There is no mention of sanitizing or escaping the content read from the addon files before the agent processes it.
  • Unverifiable Dependencies (INFO): The skill references external MCP tools (addon.deadcode, docs.stale) and related commands (c-clean, c-review) which are assumed to be part of the environment. No unauthorized remote downloads were detected.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 11:58 PM