arch-benchmark-browser SKILL.md skillA

arch-benchmark-browser from baristaze/swe_guidelines is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add baristaze/swe_guidelines/arch-benchmark-browser@git:20261007.c80a8dc

Pins git:20261007.c80a8dc in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
baristaze/swe_guidelines
Path
skills/arch-benchmark-browser/SKILL.md
Versions recorded
12
Latest version
git:20261007.c80a8dc, committed Oct 7, 2026
First version recorded
Sep 20, 2026
Size
18,482 bytes
Repository stars
94
License
MIT

Run the benchmark prompt in the chatgpt.com, claude.ai, gemini.google.com, and grok.com products, signed in, and save each answer with its conversation URL as proof; in a checkout of this repository, also check the run in, redacted, with its row.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# arch-benchmark-browser

Ask four products the same question and keep the answers with their
proof. The prompt, contract, size map, and schema live at
`../../benchmark/browser/` and
`../../benchmark/schema/browser-session.schema.json`, paths from this
skill's folder as `realpath` resolves it. If any is missing, stop and
say the installation is incomplete.

## Input

The arguments name two sizes, `model=<size> effort=<size>`, each of
`xs`, `s`, `m`, `l`, `xl`. A missing one is `m`. It may also
name a subset of sites (`sites=claude.ai,gemini.google.com`); the
default is all four. It may ask for a comparison with earlier runs:
`compare=<run_id>,<run_id>` names them, and `compare=all` means every
earlier run. Without `compare=`, nothing is compared.

## What the pages are like

Facts that decide how the steps below go. Read them before the browser.

- The read tools are permitted per domain, and not as a set. On some
  domains `get_page_text` and `read_page` are refused while a
  screenshot works; on others `find` is refused while `read_page`
  works. When a read tool is refused, use a screenshot and `zoom`, and
  transcribe from it. Say in the output which site was read that way.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add baristaze/swe_guidelines/arch-benchmark-browser@sha256:4272d7276dc2b832

The line at the top of this page pins the label git:20261007.c80a8dc, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_5j77twat63lkcwmc.svg)](https://markdownregistry.com/a/art_5j77twat63lkcwmc)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 12 versions of this file, between Sep 20, 2026 and Oct 7, 2026, about one every 2 days.

versioncommittedcommitsizeaudit
git:20261007.c80a8dc latest2026-10-07 c80a8dc 18,482 BA view · diff
git:20261007.fbc577f2026-10-07 fbc577f 18,497 BA view · diff
git:20261004.6b9120e2026-10-04 6b9120e 18,751 BA view · diff
git:20260930.c0089222026-09-30 c008922 18,404 BA view · diff
git:20260929.34220142026-09-29 3422014 18,650 BA view · diff
git:20260928.8202ef92026-09-28 8202ef9 18,656 BA view · diff
git:20260927.f6146182026-09-27 f614618 15,482 BA view · diff
git:20260922.28cefc12026-09-22 28cefc1 8,718 BA view · diff
git:20260922.e4ae97c2026-09-22 e4ae97c 8,687 BA view · diff
git:20260921.419a8e92026-09-21 419a8e9 8,937 BA view · diff
git:20260921.93b5e1e2026-09-21 93b5e1e 8,962 BA view · diff
git:20260920.65367e02026-09-20 65367e0 8,891 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (18482 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

baristaze/swe_guidelines · 94 stars · license MIT · pushed 2026-11-02 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_5j77twat63lkcwmc
GET https://markdownregistry.com/api/v1/resolve?ref=baristaze/swe_guidelines/arch-benchmark-browser
GET https://markdownregistry.com/api/v1/blob/4272d7276dc2b83203a856f0fe6a03a8c27af98cd8d1b4963f66fbf17adb5e4a

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from baristaze/swe_guidelines

CLAUDE.md@.claude CLAUDE.md file
baristaze/swe_guidelines · .claude/CLAUDE.md
git:20261004.1522cbc · audit B · 94 stars
AGENTS.md AGENTS.md file
baristaze/swe_guidelines · AGENTS.md
git:20261022.3569cc8 · audit A · 94 stars
audit-credential-lifetimes SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-credential-lifetimes/SKILL.md · Audit how long each credential keeps working after it should stop: one row per credential kind (auth session, API key…
git:20261014.62aa70e · audit B · 94 stars
audit-database-calls SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-database-calls/SKILL.md · Audit how many database calls each endpoint and each worker flow makes, at its least and at its most: round trips…
git:20261014.7748a1f · audit B · 94 stars
audit-deploy-time SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-deploy-time/SKILL.md · Audit where a deploy's minutes go: the pipeline's jobs and steps from the Actions run, the Terraform apply's resources…
git:20260929.3422014 · audit A · 94 stars
audit-provider-calls SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-provider-calls/SKILL.md · Audit every call to an external provider (WorkOS, the AWS services, any HTTP to a third party), per flow: each route…
git:20261014.7748a1f · audit A · 94 stars
audit-query-indexes SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-query-indexes/SKILL.md · Audit whether the indexes fit the queries, and what breaks first under load: every statement the Postgres storage…
git:20261014.62aa70e · audit A · 94 stars
audit-retention SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/audit-retention/SKILL.md · Audit which stores grow without bound and what trims them: every table of every role, with what writes it, when a row…
git:20261014.62aa70e · audit A · 94 stars
docs-compact SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/docs-compact/SKILL.md · Compact the tree's documents to what holds at the head of the main branch: rewrite each ADR to its present decision…
git:20261015.a329574 · audit A · 94 stars
ops-cloud-deployment-create SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/ops-cloud-deployment-create/SKILL.md · Create one cloud environment of the system from nothing, in its own AWS account, as that account's administrator: the…
git:20261014.62aa70e · audit A · 94 stars
ops-cloud-deployment-nuke SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/ops-cloud-deployment-nuke/SKILL.md · Destroy one cloud environment of the system as its account's administrator: empty the buckets, destroy the environment…
git:20261014.62aa70e · audit A · 94 stars
ops-infra-as-code SKILL.md skill
baristaze/swe_guidelines · scaffold/acme_root/.agents/skills/ops-infra-as-code/SKILL.md · Write or change the system's Terraform in the shape the guideline prescribes: one module graph shared by every…
git:20261014.62aa70e · audit A · 94 stars

Every file in baristaze/swe_guidelines

Browse by kind, by grade A, or by owner.