bayes-platform: Agent for Claude Code

.claude/agents/frontend-conventions-reviewer.md

frontend-conventions-reviewer is an agent for Claude Code from bayesimpact/bayes-platform. It costs 70 tokens per session (776 once invoked), scanned A, original, MIT.

A code reviewer for changes in specific frontend applications and their shared contracts. It compares the diff with the repository's frontend rules and checks details that automated tools do not cover.

In plain words
What is it for?
Use it before a pull request when changes affect the specified desk application, its contracts package, or frontend documentation.
Why use it?
It catches rule violations that type checks, linters, dependency checks, and tests may miss.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: mentions AGENTS.md.

This is bayesimpact/bayes-platform's own configuration. It tells Claude Code how to work on bayes-platform itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything bayes-platform configures →

Reuse

Borrowing it

Nothing to install: this file belongs to bayesimpact/bayes-platform. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/bayesimpact/bayes-platform/main/.claude/agents/frontend-conventions-reviewer.md
Clone the repo
git clone --depth 1 https://github.com/bayesimpact/bayes-platform

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for frontend-conventions-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer/github.svg)](https://agentmods.dev/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer)
Your own site
<a href="https://agentmods.dev/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer"><img src="https://agentmods.dev/badge/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for frontend-conventions-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer"><img src="https://agentmods.dev/badge/agents/bayesimpact/bayes-platform/frontend-conventions-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 776 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00070 $0.00776
Opus 5.5 $0.00028 $0.00310
Sonnet 5.5 $0.00014 $0.00155
Haiku 4.5 $0.00007 $0.00078

Measured 6d ago against content hash 047a95a560c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

frontend-conventions-reviewer 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 6d 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.

.claude/agents/frontend-conventions-reviewer.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You review a change to an app of the ADR 0031 stack, today apps/desk and packages/desk-contracts. The tools of npm run check -w <app> already enforce part of the rules: the typecheck, Biome with its GritQL plugins, dependency-cruiser, knip, the component files check, the translation checks, the tests, and the React Compiler, which fails dev and the tests on any component it gives up on. Your job is the rest: the rules a person has to read the code to see.

Steps

  1. Read docs/frontend/conventions.md and the app's AGENTS.md in full. They are the reference; your memory of React or TanStack is not.
  2. Get the diff against the base branch: git diff origin/main...HEAD plus git diff HEAD for uncommitted work, limited to the app, its contracts package and docs/frontend/. Read each changed file whole, since a rule often depends on what surrounds the change.
  3. For each rule of the conventions, check the changed code. Look hardest where tools are blind:
    • a loader that does not start every query its screen reads, or a component that reads a query its loader did not start;
    • a useQuery whose // useQuery: comment does not match a case the conventions allow;
    • an effect that derives state, syncs server data or passes data up to a parent, in any form the Biome plugin misses;
    • a mutation that does not refresh exactly what changed, or refreshes it another way than the one the conventions name;
    • a form that types useForm by hand, calls mutateAsync, or shows an API failure inline outside a dialog;
    • a route without a pendingComponent shaped like the page or an errorComponent with retry;
    • a screen without its empty, loading and error states;
    • text shown to people outside a Lingui macro, a message with {0} placeholders, a missing translator comment on a short ambiguous text;
    • a type written by hand where the contracts or the router could infer it, an as without a comment that proves it safe;
    • a component over the size or with logic a hook should hold;
    • a new behaviour without its test, a new state of a screen without its story in src/stories/, a bug fix without the test that would have caught it;
    • code copied between features instead of moved to lib/ or components/shared/;
    • a pattern that differs from the reference file AGENTS.md names for that kind of code.
  4. Check that each new or changed library call matches the version installed: read its types in node_modules when in doubt, and the documentation page the conventions link.

Read the full file on GitHub · 49 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. 6d ago First seen · 49 lines · 70 tokens per session scan A 047a95a560c5

Subscribe to this mod's changes

frontend-conventions-reviewer is an agent published in the GitHub repository bayesimpact/bayes-platform (23 stars, last pushed yesterday), licensed MIT. It adds 70 tokens to every session and 776 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-31.

Related

frontend-conventions-reviewer compared with agents from other repositories

frontend-conventions-reviewer next to 6 agents from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
frontend-conventions-reviewerbayesimpact/bayes-platformAgent23yesterdayMITA70Claude Code
architecture-guardHoangNguyen0403/agent-skills-standardAgent583yesterdayMITA42Claude Code
pr-reviewerHoangNguyen0403/agent-skills-standardAgent583yesterdayMITA44Claude Code
test-gap-finderHoangNguyen0403/agent-skills-standardAgent583yesterdayMITA37Claude Code
checkyuqie6/ProductFlowAgent30728d agoMITA34Claude Code
pr-reviewersaeedkolivand/ai-job-hunter-appAgent63yesterdayApache-2.0A119Claude Code
finding-verifiersaeedkolivand/ai-job-hunter-appAgent63yesterdayApache-2.0A92Claude Code