boxel: Skill for Claude Code

.claude/skills/bxl-engine/SKILL.md

bxl-engine is a skill for Claude Code from cardstack/boxel. It costs 163 tokens per session (5,009 once invoked), scanned A, original, MIT.

A guide to the BXL expression engine, which turns readable formulas into values and connects them to built-in functions. It covers the engine's parser, formula libraries, permissions, and changes that can silently alter results.

In plain words
What is it for?
Use it when modifying formula parsing, built-in functions, formatting, authorization, mutation handling, or test coverage in the BXL package.
Why use it?
It helps developers change the formula engine without breaking behavior that ordinary tests may not detect.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is cardstack/boxel's own configuration. It tells Claude Code how to work on boxel itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything boxel configures →

Reuse

Borrowing it

Nothing to install: this file belongs to cardstack/boxel. 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/cardstack/boxel/main/.claude/skills/bxl-engine/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/cardstack/boxel

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 bxl-engine

README.md
[![agentmods](https://agentmods.dev/badge/skills/cardstack/boxel/bxl-engine/github.svg)](https://agentmods.dev/skills/cardstack/boxel/bxl-engine)
Your own site
<a href="https://agentmods.dev/skills/cardstack/boxel/bxl-engine"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/bxl-engine/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 bxl-engine

Your own site · 80×15
<a href="https://agentmods.dev/skills/cardstack/boxel/bxl-engine"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/bxl-engine.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 163 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,009 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 $0.00163 $0.05009
Opus 5.5 $0.00065 $0.02004
Sonnet 5.5 $0.00033 $0.01002
Haiku 4.5 $0.00016 $0.00501

Measured 5d ago against content hash 715c48aa25bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

bxl-engine 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 5d 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/skills/bxl-engine/SKILL.md · 367 lines

How it starts

The opening of the file, as written. The whole thing — 367 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BXL engine — architecture and the invariants a change must preserve

packages/bxl is the expression engine: a vendored jq runtime (src/jqtools/), a vendored Excel formula library (src/formulajs/), and BXL's own readable syntax, linter, formatter, registry bridge, authorization and mutation subsystems (src/bxl/, src/authorization/, src/mutation/). It ships raw erasable TypeScript with .ts import specifiers — Node runs the sources directly, the host bundles them, and there is no build step to go stale.

This skill is for people changing the engine. Card authors writing formulas want packages/bxl/README.md and packages/bxl/docs/ instead — the syntax surface, the tag choice, the profile constraints as an author meets them.

Several of this package's invariants are enforced by the suite; the ones worth a skill are the ones that are not, because their violation leaves every test green while the thing under test quietly changes.

How a formula becomes a value

source ── compileReadableSyntax ──▶ canonical jq text     src/bxl/compiler/readable-syntax.ts
       (skipped when readableSyntax: false — the jq tag)
  │  Tokenizer → Parser                                   src/jqtools/parser/
  ▼
AST
  │  annotateBuiltinFilters      resolve each call against the registry
  │  compileScalarExpression     try to compile the WHOLE program to scalars
  ▼
either  compiledScalar(input)        one call, no jq machinery   src/jqtools/evaluate/compiledScalar.ts
or      evaluateWithRegistry(ast)    streaming Item iterators     src/jqtools/evaluate/evaluate.ts
  │  outputs unwrapped, then recordRuntimeOutput
  ▼
outputs[]

annotateBuiltinFilters (src/bxl/bridge/native.ts) walks the AST once and stamps each filter node with the implementation it resolved: resolvedJq if the registry has a jq-source definition at that NAME/arity, otherwise resolvedNative. A jq definition at a key permanently hides a native at the same key — in the annotator and in the evaluator both. The same pass marks each node singleOutput, which is what lets the evaluator evaluate an argument without materializing a stream.

Read the full file on GitHub · 367 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. 5d ago First seen · 367 lines · 163 tokens per session scan A 715c48aa25bf

Subscribe to this mod's changes

bxl-engine is a skill published in the GitHub repository cardstack/boxel (27 stars, last pushed today), licensed MIT. It adds 163 tokens to every session and 5,009 once invoked, about $0.0007 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 2026-10-21.