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.
npx agentmods add instructions/jumpstarter-dev/jumpstarter/claude-mdgit clone --depth 1 https://github.com/jumpstarter-dev/jumpstarterWhat 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 | $0.00608 | $0.00608 |
| Opus 5 | $0.00304 | $0.00304 |
| Sonnet 5 | $0.00122 | $0.00122 |
| Haiku 4.5 | $0.00061 | $0.00061 |
Grade A, and why
jumpstarter 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude AI Instructions
This file provides instructions for Claude AI when working with the Jumpstarter project.
Project Rules and Guidelines
Important project-specific rules and guidelines are located in the .claude/rules directory:
-
.claude/rules/project-structure.md: Understanding the monorepo structure, workspace configuration, package organization, and development workflows. Read this to understand how the project is organized and where files should be located. -
.claude/rules/creating-new-drivers.md: Guidelines for creating new driver packages, including naming conventions, required information, and the driver creation process. Read this when tasked with creating or modifying drivers. -
.claude/rules/releasing-operator.md: Step-by-step guide for releasing a new version of the Jumpstarter operator, including which files to update, image tag conventions, bundle generation, and community-operators contribution workflow. -
.claude/rules/jep-process.md: Process for creating Jumpstarter Enhancement Proposals (JEPs), including when to use them, numbering conventions, required sections, and the design decision format. Read this when proposing or reviewing cross-cutting changes or features that require community consensus. -
.claude/rules/e2e-doc-sync.md: Rules for keepinge2e/README.mdaligned with E2E tests, runners, and CI labels (mirrors.cursor/rules/e2e-doc-sync.mdc). Read this when changing E2E behavior or E2E documentation.
When to Read These Rules
- Always: Read
project-structure.mdwhen working with files, packages, or understanding the codebase layout - When creating drivers: Read
creating-new-drivers.mdbefore creating, improving, or documenting driver packages - When releasing the operator: Read
releasing-operator.mdbefore preparing a new operator version for OLM - When creating JEPs: Read
jep-process.mdbefore proposing enhancements that affect multiple components, change public APIs, or require community discussion - When modifying E2E tests or E2E docs: Read
e2e-doc-sync.mdand keepe2e/README.mdsynchronized in the same PR - When modifying structure: Consult both files when making changes that affect project organization
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.
- 2d ago First seen · 38 lines · 608 tokens per session scan A c2eac0ba1039
jumpstarter CLAUDE.md is an instructions file published in the GitHub repository jumpstarter-dev/jumpstarter (214 stars, last pushed 3d ago), licensed Apache-2.0. It adds 608 tokens to every session, about $0.0030 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.
Other instructions, from other repositories
allo AGENTS.md
Instructions for cornell-zhang/allo, covering building, testing, code style and don'ts.
AgentDeck CLAUDE.md
Instructions for puritysb/AgentDeck, covering agentdeck, monorepo, build, android build and setup & distribution.
rtl-agent-team CLAUDE.md
Instructions for babyworm/rtl-agent-team, covering language rule, important — project identity, why agentic coding for silicon ip, plugin runtime vs. development context (critical) and plugin architecture: dynamic prompt injection.
agent-deck AGENTS.md
Instructions for breakstring/agent-deck, covering agents.md, 项目定位, 必读文档, 工作流 and 当前代码结构.
step.parts AGENTS.md
Instructions for earthtojake/step.parts, covering agent guide, quick commands, repo shape, adding parts and metadata rules.
AgentDeck AGENTS.md
Instructions for puritysb/AgentDeck, covering agents.md, 프로젝트 지침 (antigravity & ai agents), 0. 지원 에이전트 / supported agents, 1. 컨텍스트 및 아키텍처 파악 (필수) and 2. 워크플로우 자동화 활용.