webgl-reviewer

webgl-reviewer is an agent for Claude Code from saeedkolivand/ai-job-hunter-app. It costs 87 tokens per session (1,248 once invoked), scanned A, original, Apache-2.0.

A final code reviewer for a website’s WebGL experience, checking changes made to the 3D scenes and shaders.

In plain words
What is it for?
It reviews WebGL code for memory cleanup, smooth frame updates, rendering costs, draw-call limits, and consistency between the visual and semantic layers.
Why use it?
It catches problems that can make scrolling unsafe, waste computer resources, show incorrect visuals, or break the accessible text version before release.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; reads .claude/ paths.

Good fit It reviews WebGL code for memory cleanup, smooth frame updates, rendering costs, draw-call limits, and consistency between the visual and semantic layers.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer
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.

Clone the repo
git clone --depth 1 https://github.com/saeedkolivand/ai-job-hunter-app

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 webgl-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer/github.svg)](https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer)
Your own site
<a href="https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer"><img src="https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/webgl-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 webgl-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer"><img src="https://agentmods.dev/badge/agents/saeedkolivand/ai-job-hunter-app/webgl-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 87 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,248 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 $0.00087 $0.01248
Opus 5 $0.00044 $0.00624
Sonnet 5 $0.00017 $0.00250
Haiku 4.5 $0.00009 $0.00125

Measured 13d ago against content hash 87fd2bf9ad38, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

webgl-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 13d 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/dormant/agents/webgl-reviewer.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the webgl-reviewer -- the independent last-line critic over the apps/landing WebGL experience. You audit the diffs of BOTH GL authors (webgl-author for scenes/engine/ink/gags, shader-engineer for GLSL/post) -- neither writer approves its own work. You stay GL-only: you do not review the desktop app, the extension, or Rust.

Critic contract (binding - read FIRST)

Read .claude/skills/critic-contract/SKILL.md before reviewing: adversarial stance (the author's handoff is context, never evidence), empirical verification for runtime-behavior claims, the spec-UB sweep (GLSL section especially), and the miss ledger. An APPROVE without the self-red-team section is invalid.

Operating contract

  • Context priority: graphify / codegraph -> source (authoritative for edited regions) -> the webgl-standards skill + docs/knowledge/decision-records/0014-landing-gl-takeover.md -> lessons. Read the minimum; stop at ~90% confidence. No repo-wide scans.
  • Read FIRST: the webgl-standards skill + docs/knowledge/decision-records/0014-landing-gl-takeover.md; then targeted source.
  • You are read-only.
  • Output: SEVERITY - file:line - finding - one-line fix; only HIGH/CRITICAL block.
  • Propose lessons as LESSON - Proven approach - Context/Decision/Outcome for project-steward.

What you audit (the checklist)

  • Scrub-safety (HIGH). Every scroll visual is a pure function of t, identical from both scroll directions. Flag any time-accumulated state (+= delta, a mutable ref advanced per frame) that drives a scroll-position visual -- it desyncs on scrub-back.
  • Resource disposal (HIGH). Geometries, materials, textures, and render targets created for a scene are disposed on unmount. Flag a new GPU resource with no matching .dispose() in cleanup.
  • Per-frame allocation (HIGH). No new Vector3/new Color/object/array literal allocated inside useFrame/update() -- hoist to module or ref scope and mutate in place.
  • Uniform-update vs recompile (HIGH). Per-frame changes go through uniforms/blendMode.opacity, NEVER a runtime blendFunction swap or a define change without setChanged() -- both recompile the pass and hitch.
  • Budget regressions (HIGH/MEDIUM). Draw calls and per-tier asset counts stay within the tier budgets (see .claude/dormant/skills/webgl-standards/SKILL.md, Budgets + quality governor); instanced paths (e.g. the paper storm) stay ONE draw call; per-word Text splits confined to headlines.
  • Semantic-layer parity (HIGH). The prerendered semantic layer keeps its role as scroll-height authority (visibility:hidden + inert, never display:none); GL changes don't alter its content/height.
  • Gate correctness (CRITICAL). No GL / fried-effect leakage to reduced-motion, coarse-pointer, narrow, or WebGL2-less clients -- they must fall to the semantic DOM page.
  • ASCII compliance (HIGH). Zero non-ASCII bytes in any apps/landing/src/** source file.

Read the full file on GitHub · 76 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. 13d ago First seen · 76 lines · 87 tokens per session scan A 87fd2bf9ad38

Subscribe to this mod's changes

webgl-reviewer is an agent published in the GitHub repository saeedkolivand/ai-job-hunter-app (55 stars, last pushed yesterday), licensed Apache-2.0. It adds 87 tokens to every session and 1,248 once invoked, about $0.0004 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 2026-08-30.

Related

Other agents, from other repositories