sniff CLAUDE.md

sniff CLAUDE.md is an instructions file for coding agents from Aboudjem/sniff. It costs 1,316 tokens per session, scanned A, original, Apache-2.0.

Project instructions for Sniff, a command-line quality checker that opens a running web app in a real browser and follows user flows. It reports broken pages and forms, accessibility issues, browser-console errors, and responsive problems with screenshots; source-only scanning is also available.

In plain words
What is it for?
Use it to scan a web app from the command line, check routes and user flows across browsers, and collect screenshot evidence for functional, visual, accessibility, and console issues.
Why use it?
It helps find problems through actual browser behavior instead of relying only on source inspection or manually written tests.

Instructions file

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 instructions/aboudjem/sniff/claude-md
Clone the repo
git clone --depth 1 https://github.com/Aboudjem/sniff

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 sniff CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/aboudjem/sniff/claude-md.svg)](https://agentmods.dev/instructions/aboudjem/sniff/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/aboudjem/sniff/claude-md"><img src="https://agentmods.dev/badge/instructions/aboudjem/sniff/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,316 This file is loaded in full into every session.
When invoked 1,316 The same file — it is already loaded in full.
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 $0.01316 $0.01316
Opus 5 $0.00658 $0.00658
Sonnet 5 $0.00263 $0.00263
Haiku 4.5 $0.00132 $0.00132

Measured today against content hash f216f6968f67, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sniff CLAUDE.md 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 today.

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.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Project

Sniff

An autonomous QA scanner that walks your running app in a real headless browser, follows real user flows, and reports what is actually broken (broken pages, forms, accessibility, console errors, responsive issues), each with screenshot proof. The flow-walk is the default (sniff or sniff --url); sniff scan is the source-only mode. Distributed as an npm CLI and Claude Code plugin, with no API key required.

Core Value: One command finds bugs across every dimension (functional, visual, accessibility, performance) before users do, with no manual test writing, ever.

Constraints

  • Tech stack: TypeScript + Node.js 22+ (locked decision)
  • Test runtime: Playwright (chromium, webkit, firefox), a locked decision
  • Default AI: Claude Code (no API key), the key differentiator, must work out of the box
  • Distribution: npm-first (npx sniff), the lowest-friction path for developers
  • Budget: Zero infrastructure cost for users, everything runs locally
  • Performance: Must complete a scan of a medium app (50 routes) in under 10 minutes

Technology Stack

Technology stack not yet documented. Will populate after codebase mapping or first phase.

Conventions

Conventions not yet established. Will populate as patterns emerge during development.

Architecture

Architecture not yet mapped. Follow existing patterns found in the codebase.

Project Skills

No project skills found. Add skills to any of: .claude/skills/, .agents/skills/, .cursor/skills/, or .github/skills/ with a SKILL.md index file.

Sniff contributor notes (portability and discoverability layer)

These notes record the gotchas introduced by the multi-CLI install and discoverability work. Keep them current when you touch the related files.

Host-agnostic agents

Sniff has no agents/ directory, so there is no model: frontmatter to strip. If agents are ever added, omit model: from their frontmatter so each host CLI falls back to its own default model. A literal model: inherit is a Claude-Code-only keyword that some hosts (for example OpenCode) reject as an unknown model id, which is why host-agnostic plugins carry only name and description.

Read the full file on GitHub · 96 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. today Changed · -5 lines · +71 tokens per session f216f6968f67
  2. 4d ago First seen · 101 lines · 1,245 tokens per session scan A 3559e6cf2a5a

Subscribe to this mod's changes

sniff CLAUDE.md is an instructions file published in the GitHub repository Aboudjem/sniff (4 stars, last pushed today), licensed Apache-2.0. It adds 1,316 tokens to every session, about $0.0066 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.