What is skills-benchmarks CLAUDE.md?

A set of project rules for LangSmith skills, which are reusable instructions for coding agents, in both Python and TypeScript.

Use it when changing skill scripts, updating skill Markdown files, or checking that Python and TypeScript implementations match.

It helps keep the two language versions, their commands, outputs, errors, and documentation aligned.

How langchain-ai/skills-benchmarks configures Claude Code

About the project

skills-benchmarks is a test suite that measures how the design of skill documentation affects Claude Code's adherence to recommended coding patterns. It is used to compare documentation approaches across LangChain-related tasks and other agent workflows. Its catalogue entries represent skills, hooks, instructions, and a plugin used in the benchmark project.

langchain-ai/skills-benchmarks · 116 stars · on GitHub

Reuse

Borrowing it

Nothing to install: this file belongs to langchain-ai/skills-benchmarks. 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/langchain-ai/skills-benchmarks/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/langchain-ai/skills-benchmarks

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/langchain-ai/skills-benchmarks/claude-md/github.svg)](https://agentmods.dev/instructions/langchain-ai/skills-benchmarks/claude-md)

Your own site

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

Per session 1,324 This file is loaded in full into every session.

When invoked 1,324 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	$0.01324	$0.01324
Opus 5	$0.00662	$0.00662
Sonnet 5	$0.00265	$0.00265
Haiku 4.5	$0.00132	$0.00132

Measured 9d ago against content hash 32dabac8d33b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

skills-benchmarks 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 9d 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 · 148 lines

How it starts

The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skills Project Guidelines

Python/TypeScript Parity

CRITICAL: LangSmith skills have both Python and TypeScript implementations. These MUST stay in sync:

Same CLI commands, flags, and options
Same output format for identical inputs
Same error handling behavior

When modifying any script, always update both Python and TypeScript versions together. Parity tests in tests/scripts/langsmith/parity/ verify this.

The TypeScript scaffold (scaffold/typescript/) mirrors Python:

Both scaffolds are copied into Docker so test scripts in either language work from either runner
TestRunner class available in both languages with identical API
buildTreatmentSkills() in TS doesn't support included_sections, section_overrides, extra_sections — use Python for full section manipulation

Skill Markdown File Parity

Each skill directory contains variant files that MUST stay aligned:

skill_py.md - Python-only content
skill_ts.md - TypeScript/JavaScript-only content
skill_all.md - Combined content for both languages

When updating skill documentation:

All three files should have the same section structure (headers, order)
Code examples should be equivalent implementations in each language
CLI commands should show the appropriate language-specific invocation
Keep descriptions consistent across variants

Task Structure

Each task is a self-contained directory:

tasks/my-task/
  task.toml             # Metadata + validation config
  instruction.md        # Task prompt with {variable} placeholders
  environment/          # Docker context (Dockerfile, requirements.txt, source code)
  validation/           # Test scripts (run inside Docker)
  data/                 # Ground truth, test cases (optional)

Validation is config-driven via task.toml:

[validation]
test_scripts = "test_my_task.py"       # Script(s) to run in Docker
target_artifacts = ["output.py"]       # File(s) Claude should create
timeout = 120                          # Docker execution timeout

Read the full file on GitHub · 148 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.

9d ago First seen · 148 lines · 1,324 tokens per session scan A 32dabac8d33b

Subscribe to this mod's changes

skills-benchmarks CLAUDE.md is an instructions file published in the GitHub repository langchain-ai/skills-benchmarks (116 stars, last pushed 21d ago), licensed MIT. It adds 1,324 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 2026-08-30.

Testing

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

skills-benchmarks: Instructions file for Claude Code