semantic-markup-contracts

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides utility shell commands using the rg (ripgrep) tool to perform static analysis on the local filesystem. These commands are restricted to read-only search operations within project directories (src/, app/, packages/) to identify accessibility smells like improper onClick usage on non-interactive elements or redundant ARIA roles.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and review source code, which creates a potential surface for indirect prompt injection if the code being analyzed contains adversarial patterns designed to influence the agent's behavior.
  • Ingestion points: Local source code files in src/, app/, and packages/ (SKILL.md).
  • Boundary markers: None explicitly defined in the instruction text to delimit external code from agent instructions.
  • Capability inventory: Use of shell search tools (rg) and providing code refactoring recommendations.
  • Sanitization: The instructions do not specify a process for sanitizing or escaping the content of the files being audited.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 11:59 PM
Security Audit — agent-trust-hub — semantic-markup-contracts