python-lsp CLAUDE.md

Project instructions for zircote/python-lsp, a Claude Code plugin that supports Python development through Pyright, a Python type checker, and automated hooks. The hooks can format, lint, type-check, test, and scan Python code and dependencies for security issues.

In plain words
What is it for?
Use it when modifying the plugin, setting up its development tools, or working on its language-server configuration and automated hook scripts.
Why use it?
It explains how to set up the project and what runs after files change, so an agent can work within its existing checks and tools.

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/zircote/python-lsp/claude-md
Clone the repo
git clone --depth 1 https://github.com/zircote/python-lsp
Per session 454 This file is loaded in full into every session.
When invoked 454 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 $0.00454 $0.00454
Opus 5 $0.00227 $0.00227
Sonnet 5 $0.00091 $0.00091
Haiku 4.5 $0.00045 $0.00045

Measured yesterday against content hash a58a2ddc3bd7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

python-lsp 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 yesterday.

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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

CLAUDE.md · 55 lines

What it actually says

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

A Claude Code plugin providing Python development support through Pyright LSP integration and 15 automated hooks for type checking, linting, formatting, testing, and security scanning.

Setup

Run /setup to install all required tools, or manually:

npm install -g pyright
pip install ruff black isort mypy bandit pytest pip-audit

Key Files

File Purpose
.lsp.json Pyright LSP configuration
hooks/hooks.json Automated development hooks
hooks/scripts/python-hooks.sh Hook dispatcher script
commands/setup.md /setup command definition
.claude-plugin/plugin.json Plugin metadata

Hook System

All hooks trigger afterWrite. Hooks use command -v checks to skip gracefully when optional tools aren't installed.

Hook categories:

  • Formatting (**/*.py): black/ruff format, isort
  • Linting (**/*.py): ruff/flake8, type checking
  • Security (**/*.py): bandit, pip-audit
  • Dependencies (**/requirements.txt, **/pyproject.toml): audit, validation

When Modifying Hooks

Edit hooks/scripts/python-hooks.sh. The script handles different file types and runs appropriate tools:

  • Use || true to prevent hook failures from blocking writes
  • Use head -N to limit output verbosity
  • Use command -v tool >/dev/null && for optional tool dependencies

When Modifying LSP Config

Edit .lsp.json. The extensionToLanguage map controls which files use the LSP. Current config maps .py and .pyi files to the python language server.

Conventions

  • Prefer minimal diffs
  • Keep hooks fast (use --quiet, limit output with head)
  • Documentation changes: update both README.md and commands/setup.md if relevant
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. yesterday First seen · 55 lines · 454 tokens per session scan A a58a2ddc3bd7

Subscribe to this mod's changes

python-lsp CLAUDE.md is an instructions file published in the GitHub repository zircote/python-lsp (3 stars, last pushed 22d ago), licensed MIT. It adds 454 tokens to every session, about $0.0023 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.