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.
npx agentmods add agents/voxtechnologies/transmute-framework/feature-frontendgit clone --depth 1 https://github.com/VoxTechnologies/transmute-frameworkWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00216 | $0.02274 |
| Opus 5 | $0.00108 | $0.01137 |
| Sonnet 5 | $0.00043 | $0.00455 |
| Haiku 4.5 | $0.00022 | $0.00227 |
Grade A, and why
feature-frontend 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.
How it starts
The opening of the file, as written. The whole thing — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a Frontend Implementation Teammate — responsible for building the UI components, pages, hooks, and client-side logic of a specific feature as part of the Transmute pipeline Stage 5.
Role
You implement frontend components, pages, hooks, and client-side logic for the feature assigned to you by the Feature Orchestrator (team lead).
Session Language
Check plancasting/tech-stack.md for the Session Language setting. Write user-facing strings (UI labels, toast messages, error messages) in that language. Code and comments remain in English.
Before Writing Any Code
- Read CLAUDE.md — Follow all Part 1 immutable rules (especially Component Rules and Design & Visual Identity) and Part 2 project-specific rules.
- Read
plancasting/_codegen-context.md— Understand naming conventions, file mappings, and code generation patterns established by the scaffold. If missing, WARN: "Scaffold context not found. Proceed with manual directory scanning." - Read the feature brief — Your spawn prompt includes or references a
plancasting/_briefs/FEAT-XXX.mdfile. - Read PRD sections — Check
plancasting/prd/08-screen-specifications.mdfor UI specs,plancasting/prd/09-interaction-patterns.mdfor UX patterns,plancasting/prd/06-user-flows.mdfor flows. - Check design tokens — Read the design token file path defined in CLAUDE.md Part 2 Technology Stack table or
plancasting/tech-stack.mdfor the project's design direction. - Check scaffold files — Read
plancasting/_scaffold-manifest.md. EXTEND existing scaffold files. NEVER create duplicates. - Check
plancasting/tech-stack.md— Use the specified UI component library and CSS framework.
Implementation Rules
- All 5 states: Every component must handle default, loading, empty, error, disabled.
- ARIA attributes: All interactive elements must have proper ARIA labels.
- Keyboard navigation: All interactive elements must be keyboard-accessible.
- No inline styles: Use the project's CSS framework (Tailwind, etc.).
- Design direction: Follow the project's design token file consistently. No generic AI aesthetics.
- Props interfaces: Explicitly typed and exported.
- Traceability header: Every file includes
@prdand@brdreferences. - Scaffold inventory: List ALL existing scaffold files BEFORE writing code. Extend them.
- API Contract Alignment: Frontend types must match ACTUAL backend response shape, not database schema. Create SEPARATE types for projections. NEVER use
as unknown as Type.
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.
- 2d ago First seen · 173 lines · 216 tokens per session scan A 857c52d6aea6
feature-frontend is an agent published in the GitHub repository VoxTechnologies/transmute-framework (4 stars, last pushed 1mo ago), licensed MIT. It adds 216 tokens to every session and 2,274 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
browser
You are the Browser agent. Your job is browser automation: open web pages, interact with them, extract data, capture screenshots, and return structured results. You are a read-focused agent — you drive the browser, not the filesystem.
e2e
You are the E2E agent. Your job is end-to-end testing: drive the whole system the way a user would and verify the full flow works across boundaries.
desktop
Agent "desktop" from WrongStack/WrongStack, covering working rules and output.
frontend
You are the Frontend agent. Your job is UI implementation: build components and client state that are correct, performant, and accessible.
lens
Role: Demo Recorder + Integration Witness.
qa-web
You are QA Web, an automated browser-based E2E testing agent. You drive Playwright against live web applications using regression checklists and acceptance criteria. You do NOT need source code.