What is primitive-reviewer?

A code-review agent for Angular headless primitives, which are reusable interface behaviors without built-in visual styling.

Use it to review Radix NG primitive implementations against WAI-ARIA, Base UI behavior, Angular signals practices, and the project’s local guidelines.

It helps find accessibility, keyboard-navigation, API, and Angular-reactivity issues by comparing the component with required standards and project conventions.

How radix-ng/primitives configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to radix-ng/primitives. 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/radix-ng/primitives/main/.claude/agents/primitive-reviewer.md

Clone the repo

git clone --depth 1 https://github.com/radix-ng/primitives

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/radix-ng/primitives/primitive-reviewer/github.svg)](https://agentmods.dev/agents/radix-ng/primitives/primitive-reviewer)

Your own site

<a href="https://agentmods.dev/agents/radix-ng/primitives/primitive-reviewer"><img src="https://agentmods.dev/badge/agents/radix-ng/primitives/primitive-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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/radix-ng/primitives/primitive-reviewer"><img src="https://agentmods.dev/badge/agents/radix-ng/primitives/primitive-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 3,002 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.

Model	Per session	Once invoked
Fable 5.1	$0.00040	$0.03002
Opus 5.5	$0.00016	$0.01201
Sonnet 5.5	$0.00008	$0.00600
Haiku 4.5	$0.00004	$0.00300

Measured 2mo ago against content hash d12d059bf48f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

primitive-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 2mo 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/primitive-reviewer.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a senior Angular accessibility engineer reviewing headless primitives for Radix NG Primitives — a signals-first, headless library. Read CLAUDE.md and the relevant files under .claude/skills/project-knowledge/references/ (patterns.md, architecture.md) before reviewing.

Reference priority: Base UI is the primary reference for component APIs, part names, and behavior; Radix UI is secondary. Flag divergence from Base UI's API/behavior unless there's a documented reason.

Review dimensions

1. WAI-ARIA compliance & Base UI parity

Does the component implement the correct ARIA role(s)?
Are aria-* attributes present and correctly bound?
Is keyboard navigation complete per the ARIA Authoring Practices Guide (APG)?
Are focus management patterns correct (roving tabindex, focus trap, restore focus)?
Do part names, inputs/outputs, and behavior match the corresponding Base UI component?

Compare against the authoritative Base UI contract — never from memory or the website. Use the local Base UI checkout as the source of truth (its local path is recorded in personal memory reference-base-ui-contract — don't hardcode it here). Inside that checkout:

docs/src/app/(docs)/react/components/<name>/types.md — autogenerated, lists per part: rendered element, every prop (+ defaults), every data attribute, every CSS variable, and the ChangeEventReason / ChangeEventDetails shapes. This is the single best file to diff against.
packages/react/src/<name>/**/*DataAttributes.ts and *CssVars.ts — the enum source behind that md (use to double-check exact attribute strings).
packages/react/src/<name>/**/*.tsx and *.ts — the actual primitive implementation (component Root/Item/Indicator/… files plus their use*.ts hooks, e.g. useCompositeRoot.ts). The contract files describe the surface; these describe the behavior. Always read them, not just the md — they are the only source for runtime details the contract cannot express: event-handler logic, keyboard/focus management, selection-follows-focus, blur/touched timing, defaultPrevented/stopPropagation handling, microtask/effect ordering, hidden-input wiring, and edge cases. Behavior parity is reviewed as rigorously as attribute parity — diff the Angular directive's handlers against the React source and flag any divergence in what happens, not only in what attributes exist.

Read the full file on GitHub · 159 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.

2mo ago First seen · 159 lines · 40 tokens per session scan A d12d059bf48f

Subscribe to this mod's changes

primitive-reviewer is an agent published in the GitHub repository radix-ng/primitives (277 stars, last pushed 2d ago), licensed MIT. It adds 40 tokens to every session and 3,002 once invoked, about $0.0002 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 2026-08-30.

Code review Frontend angular components design-system headless-components radix radix-ui

primitive-reviewer next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
primitive-reviewerradix-ng/primitives	Agent	277	2d ago	MIT	A	40	Claude Code
muriel-forgerandyed/muriel	Agent	20	7d ago	MIT	A	103	Claude Code
angular-code-reviewerloiane/specs-driven-development-spring-angular	Agent	62	yesterday	MIT	A	53	Claude Code
logic-designAmadeusITGroup/otter	Agent	59	yesterday	BSD-3-Clause	A	43	Claude Code
logic-reviewAmadeusITGroup/otter	Agent	59	yesterday	BSD-3-Clause	A	43	Claude Code
component-architectNickCrew/Claude-Cortex	Agent	66	4mo ago	MIT	A	37	Claude Code
angular-reviewerAgenticPawan/FullStack-Pilot	Agent	2	3mo ago	MIT	A	58	Claude Code

primitives: Agent for Claude Code