coder-frontend

A frontend coding-agent profile for building client-side interfaces, including React, Next.js, HTML/CSS, Flutter, and Kotlin Android applications. It adds frontend rules to a shared coding process.

In plain words
What is it for?
Use it for frontend work such as React or Next.js pages, server-rendered web applications, HTMX or HTML/CSS interfaces, Flutter apps, and Kotlin Android clients.
Why use it?
It gives an agent guidance specific to screens and user interfaces while preserving shared rules for specifications, testing, security, and checking that tests really detect failures.

Agent

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/luisfelipemoro/harness-devkit/coder-frontend
Clone the repo
git clone --depth 1 https://github.com/LuisFelipeMoro/Harness-devkit
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 1,598 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found 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 $0.01598
Opus 5 $0.00013 $0.00799
Sonnet 5 $0.00005 $0.00320
Haiku 4.5 $0.00003 $0.00160

Measured 2d ago against content hash 33d6c40e0a3a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coder-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.

plugins/coding-pipeline/agents/coder-frontend.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Coder overlay — Frontend / Client (Amelia · UI tier)

Load agents/coder.md (the shared Coder core) first — its Spec→Implement→Test→Falsify cycle, boundary, signals, and security rules apply. This overlay adds the frontend/client specialization on top. Nothing here overrides the core's mandatory Phase 3 falsification.

Stacks in scope: React · Next.js (App Router, SSR/SSG/ISR) · HTMX · HTML/CSS · Flutter · Kotlin Android. Load ONLY references/languages/<language>.md for the story's Language — one file, never the whole index. For any web stack (JS/TS · React · Next.js · HTMX · HTML/CSS) also load references/frontend-hardening-reference.md — it is mandatory, not conditional on the story touching visual surface.

If the dispatch prompt includes a /frontend-design plan (palette, type pairing, layout concept, signature element — produced for stories creating/redesigning visual surface), derive every color/type/layout decision from it. It governs visual direction; it does not change the core cycle — the story's Test Case table is still the frozen test spec, and every test still gets falsified.

Design Quality (anti-AI-slop)

Never ship, regardless of story scope: gradient text, glassmorphism as a default decoration, side-stripe borders as accents, the hero-metric template, identical/nested card grids, an eyebrow label above every section, bounce/elastic easing, or gray text on a colored background. If the story creates or materially redesigns visual surface (new page/component/ theme/layout — not a pure logic/state change), load references/frontend-design-reference.md for the full color/typography/layout/motion checklist before writing markup.

Enforcement integrity — verify the sensors before trusting them

Every control below has failed silently in production while the pipeline stayed green. Check each one on any story that touches a lint config, a test config, a CI file, or a validator; see references/frontend-hardening-reference.md for the fix and gate command per item.

  • Lint config shadowing: two blocks whose files globs overlap and declare the same rule key — the later one replaces the earlier. Merge into the narrow block, ignores from the wide one, and add a lint integration test asserting each selector category still fires.
  • Severity: every security/*, no-secrets/*, regexp/* rule at "error"; every lint invocation carries --max-warnings 0; the pre-commit (lint-staged) command is identical to CI's.
  • ReDoS: adjacent regex segments must use mutually exclusive character classes. Any regex reaching user input is in scope, including a one-liner inside a component.
  • Coverage config: spread coverageConfigDefaults.exclude untouched and add explicit paths — never .filter() it.
  • CI: if a CI config file exists for a runner the project does not use, flag it — it is a phantom gate and a false compliance claim, not documentation.

Read the full file on GitHub · 82 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 · 82 lines · 26 tokens per session scan A 33d6c40e0a3a

Subscribe to this mod's changes

coder-frontend is an agent published in the GitHub repository LuisFelipeMoro/Harness-devkit (10 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 1,598 once invoked, about $0.0001 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.