frankxai

Trust Reviewer — Data & AI agent for Claude Code

Data & AI community

Traces one retrieved document from the retriever to the context window and names the line where it becomes labelled data, assigns a trust tier to every input and tool, and lists every irreversible too.

How to install Trust Reviewer

Installs to ~/.claude/agents/frankxai-ai-architect-trust-reviewer.md

Terminal

mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/frankxai/ai-architect/HEAD/agents/trust-reviewer.md -o ~/.claude/agents/frankxai-ai-architect-trust-reviewer.md

Restart Claude Code, or start a new session, for it to be picked up.

What Trust Reviewer does

name: trust-reviewer description: Traces one retrieved document from the retriever to the context window and names the line where it becomes labelled data, assigns a trust tier to every input and tool, and lists every irreversible tool with the human step that gates it. Use for the secure stage of the AI Architect lifecycle, or when asked to "where is the trust boundary", "check this for prompt injection", "what can this agent do without asking", "which tools are irreversible", or "audit the

Alternatives in Data & AI

Info Disclosure — Information Disclosure specialist (H1 #18, CWE-200/209/215/538/668/798) 812 ★
Stack Auditor — Audits a codebase's AI agent stack against the live best-of-Agent-Harnesses dataset — finds which harnesses th 801 ★
Multi Agent Systems Architect Agent — Systems architect specializing in the design, coordination, and governance of multi-agent AI pipelines — cover 669 ★