UI Payload Builder — Security agent for Claude Code
Extracts only what the UI needs from the final research output (analysis.json, final_etf_decision.md, data_coverage.md) into ui_payload.json plus an audit log.
How to install UI Payload Builder
Installs to ~/.claude/agents/jingi723-etf-research-agent-ui-payload-builder.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/jingi723/etf-research-agent/HEAD/.claude/agents/ui-payload-builder.md -o ~/.claude/agents/jingi723-etf-research-agent-ui-payload-builder.md Restart Claude Code, or start a new session, for it to be picked up.
What UI Payload Builder does
name: ui-payload-builder description: "Extracts only what the UI needs from the final research output (analysis.json, final_etf_decision.md, data_coverage.md) into ui_payload.json plus an audit log. Pure extraction — it creates no grades and no judgments."
UI Payload Builder
You are an extractor. **You do not analyse** — you rearrange the grades, verdicts, and prose that upstream already settled into screen-sized units. You are the only point where screen data comes out of analysis.j
Alternatives in Security
- Security Requirement Extraction — Transform threat analysis into actionable security requirements 31.9k ★
- Dr Audit — Sub-agent prompt: audit Decision Records in an implementation-plan.md for canonical form and reference resolut 432 ★
- Authz Auditor — Phase 6 authorization and access-control audit agent that enumerates every route/handler/consumer across the c 125 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Corp Auditor
A /corp AUDIT seat on Sonnet at high effort: claim audit, extraction, reconciliation, PMO. Spawn /corp audit s
Continuous Reviewer
Pinned per-task cross-reviewer for builder-pool waves. Reviews each builder's collected lease output for test/
Team Finisher
Final cleanup specialist. Removes all console.log statements (including audit diagnostic logs) and enforces co
Sec Researcher
Read-only security research for ClaudeSec — threats, CVEs, attack vectors, and compliance frameworks (NIST, IS
Code Delivery Orchestrator
Use this agent when a coding agent has finished writing code and needs to hand off the delivery process. This
Fact Reviewer
Independent fact-checker for completed analysis pipelines. Use when an analysis (model + structured output + f
Related Skills
Supabase DB Supervisor
Fresh-eyes supervision pass for AI-built database work. The builder never grades its own homework.
Invoice Auditor
A Claude skill that extracts invoices and proves the extraction is correct. Model extracts, code audits the ar
Cjm Harness Transcripts
Parse agent-harness session transcripts (Claude Code JSONL first): the append-only record DAG, derived active