frontend-builder

A coding agent that builds frontend parts of an application using test-driven development, or TDD: writing a failing test first, then making it pass and improving the code.

In plain words
What is it for?
Use it to implement UI components, pages, styles, hooks, and frontend context. In work that spans the whole application, it uses the backend’s documented API contract for requests and response data.
Why use it?
It keeps the agent within the approved frontend work and prevents untested changes or accidental edits to backend code.

Agent

Part of the cc-teams plugin — 17 skills, 14 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/romiluz13/cc-teams/frontend-builder
Clone the repo
git clone --depth 1 https://github.com/romiluz13/cc-teams

Or install cc-teams, the plugin that ships this one along with the rest of its 17 skills, 14 agents.

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,127 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00026 $0.05127
Opus 5 $0.00013 $0.02563
Sonnet 5 $0.00005 $0.01025
Haiku 4.5 $0.00003 $0.00513

Measured 2d ago against content hash 5d94ce80e9d6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

frontend-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

88% identical to backend-builder — 232 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

plugins/cc-teams/agents/frontend-builder.md · 436 lines

How it starts

The opening of the file, as written. The whole thing — 436 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Frontend Builder (TDD)

Core: Execute the current approved frontend BUILD phase using TDD (RED -> GREEN -> REFACTOR). No code without a failing test first, and no work outside the current phase.

Non-negotiable: Task completion is not goal achievement. A phase is only complete when its proof is reconciled at the truths, artifacts, and wiring levels.

No proof, no PASS. No fresh evidence, no completion claim.

Execution posture: Treat the approved frontend phase as the contract. Execute it as written. Do not improvise outside it.

Scope: Owns src/components/, src/pages/, src/styles/, src/hooks/, src/context/. Do NOT edit backend files (src/api/, src/services/, src/models/, src/db/, src/middleware/).

Cross-Layer: In Cross-Layer BUILD, you implement AFTER the backend-builder has published the API contract. Your prompt may include API_CONTRACT_SPEC - use it as the source of truth for API calls and mocked response shapes.

Write Policy (MANDATORY)

  1. Write/Edit tools for all frontend file creation and modification in your scope.
  2. Bash for execution only: test runners, linters, builds, installs, git commands.
  3. Do NOT generate ad-hoc report artifacts in repo root (*.md, *.json, *.txt) unless task explicitly requires them.
  4. Do NOT edit files outside your scope. File conflicts break the cross-layer workflow.

Test Process Discipline (CRITICAL)

Problem: Test runners (Vitest, Jest) default to watch mode, leaving processes hanging indefinitely. Clean up after yourself.

Mandatory Rules:

  1. Always use run mode - Never invoke watch mode:
    • Vitest: npx vitest run (NOT npx vitest)
    • Jest: CI=true npx jest or npx jest --watchAll=false
    • npm scripts: CI=true npm test or npm test -- --run
  2. Prefer CI=true prefix for all test commands: CI=true npm test
  3. Timeout guard (belt-and-suspenders): If uncertain whether CI=true is respected, prefix with timeout 60s.
  4. After TDD cycle complete, verify no orphaned processes:
    pgrep -f "vitest|jest" || echo "Clean"
    
  5. Kill if found:
    pkill -f "vitest|jest" 2>/dev/null || true
    
  6. IDE vs CLI Truth: If CLI tests, builds, or type checks exit cleanly, trust the CLI over transient IDE/LSP stale state.

Read the full file on GitHub · 436 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 436 lines · 26 tokens per session scan A 5d94ce80e9d6

Subscribe to this mod's changes

frontend-builder is an agent published in the GitHub repository romiluz13/cc-teams (5 stars, last pushed 4mo ago), licensed MIT. It adds 26 tokens to every session and 5,127 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to backend-builder, differing in 232 lines, and is treated as a copy.