What is bram CLAUDE.md?

A set of instructions for developing Bram, a desktop app that places a terminal beside an agent conversation pane and may show a separate web app. It explains which files belong to Bram and how to distinguish them from the target app being developed.

Use it when modifying Bram’s interface, working with its XMLUI files, or helping develop the separate target app shown in the optional pane. It also explains how to check whether that pane is available.

It prevents changes meant for the target app from being made in Bram’s own interface, or the other way around. It also clarifies that the target app may use different web technologies.

How judell/bram configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to judell/bram. 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/judell/bram/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/judell/bram

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/instructions/judell/bram/claude-md/github.svg)](https://agentmods.dev/instructions/judell/bram/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/judell/bram/claude-md"><img src="https://agentmods.dev/badge/instructions/judell/bram/claude-md/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/instructions/judell/bram/claude-md"><img src="https://agentmods.dev/badge/instructions/judell/bram/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,552 This file is loaded in full into every session.

When invoked 2,552 The same file — it is already loaded in full.

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	$0.02552	$0.02552
Opus 5.5	$0.01021	$0.01021
Sonnet 5.5	$0.00510	$0.00510
Haiku 4.5	$0.00255	$0.00255

Measured 2mo ago against content hash 6c7bd0570d97, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

bram 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 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.md · 207 lines

How it starts

The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bram

You are running in the terminal of a Tauri desktop shell. The shell puts a real terminal (where you run) next to an agent pane the user can SEE while talking to you. It can optionally also show a right-pane target-app iframe, but that pane is off by default and often absent — most users run their app in their own browser. Detect before you assume an iframe is there; when one is present, use it.

Keep two distinct surfaces straight — they are not the same, and the rules differ:

The agent pane — Bram's own UI (the Worklist / Transcript / Sessions / Context / Status tabs). Always XMLUI — that's how Bram is built. Editing it means app/tools/Main.xmlui, app/tools/components/*.xmlui, app/__shell/helpers.js, app/tools/Globals.xs.
The target app — whatever project the user is developing with Bram's help, shown in the optional target-app iframe when that pane is enabled (off by default, often absent). Per app/__shell/conventions.md, Bram "works with any project that serves a web UI (vanilla HTML/JS, a React or other Node app, a Python web app, an XMLUI app, etc.)." It may or may not be XMLUI — detect before you assume, and don't assume the pane is even present.

The XMLUI-specific guidance below is unconditional when working on Bram itself, and applies to the target app only when the target is XMLUI.

Working on Bram itself (the agent pane is XMLUI)

Most edits to Bram land in .xmlui files. Rules the xmlui-standalone evaluator enforces hard:

No raw browser JS in event handlers — setTimeout, setInterval, fetch outside DataSource, async / await, etc. are rejected at evaluation time with an unhandled rejection. Stay within App abstractions: delay(ms), debounce(ms, fn, ...args), the Timer component, DataSource for HTTP, ChangeListener for derived reactivity.
Lead with the xmlui-mcp tools before reaching for a JS solution. The xmlui_search_howto tool is the fastest way to find the XMLUI-native pattern for a feature (e.g. "delay function", "debounce input", "wrap text in table cell"); xmlui_component_docs is for component-prop lookups; xmlui_get_prompt re-injects the server's framing guidance mid-session when you suspect you've drifted.
Cite a doc URL for any non-obvious markup decision — https://www.xmlui.org/docs/reference/components/<Name> or https://www.xmlui.org/docs/howto/<slug>. If you can't cite one, search again.

Read the full file on GitHub · 207 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 · 207 lines · 2,552 tokens per session scan A 6c7bd0570d97

Subscribe to this mod's changes

bram CLAUDE.md is an instructions file published in the GitHub repository judell/bram (57 stars, last pushed yesterday), licensed Apache-2.0. It adds 2,552 tokens to every session, about $0.0102 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.

Building agents Frontend

bram CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
bram CLAUDE.mdjudell/bram	Instructions file	57	yesterday	Apache-2.0	A	2,552	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

bram: Instructions file for Claude Code