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/piratuks/invoice-builder/copilot-instructionsgit clone --depth 1 https://github.com/piratuks/invoice-builderWrote 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.
[](https://agentmods.dev/instructions/piratuks/invoice-builder/copilot-instructions)<a href="https://agentmods.dev/instructions/piratuks/invoice-builder/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/piratuks/invoice-builder/copilot-instructions.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01320 | $0.01320 |
| Opus 5 | $0.00660 | $0.00660 |
| Sonnet 5 | $0.00264 | $0.00264 |
| Haiku 4.5 | $0.00132 | $0.00132 |
Grade A, and why
invoice-builder copilot-instructions.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 4d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot instructions for Invoice Builder
Overview
This repository is an Electron app with a separate backend webserver used for the web/docker mode. Key runtime pieces are:
- Electron main process: built by
viteviavite.electron.config.ts-> outputsdist-be/backend/main/main.cjs - Preload script:
src/preload/preload.ts-> built withvite.preload.config.ts-> outputsdist-be/preload/preload.cjs - Renderer (React + Vite):
src/renderer(dev:vite, build:vite build) - Optional backend webserver:
src/backend/webserver/main.ts(dev:npm run dev:webserver, build:npm run build:webserver) - Migrations: source files in
src/backend/shared/migrations, built intodist-be/backend/migrationsusingvite.migrations.config.ts.
BMAD-inspired workflow for agentic development
This repository uses a lightweight BMAD-style workflow for coding tasks. See AGENTS.md for the complete workflow guide, including when to use this workflow, default verification checks, and team expectations.
For new work with the agent-based approach:
- Use the
poagent with bmad-po.prompt.md for acceptance criteria and scope clarification. - Use the
devagent with bmad-dev.prompt.md for implementation. - Use the
qaagent with bmad-qa.prompt.md for review and verification. - Use the
deliveryagent with bmad-delivery.prompt.md for slicing and sequencing release work. - Use the
scrumagent with bmad-scrum.prompt.md for sprint planning and execution cadence.
Repository-specific expectations:
- UI work usually touches
src/renderer. - IPC or preload changes require matching updates in
src/preload/preload.tsand the Electron main process undersrc/backend/main. - Database or persistence changes should be reviewed for migrations under
src/backend/shared/migrations. - Webserver changes should be validated through the appropriate build or dev workflow.
Primary developer commands
- Full local dev (electron + renderer + preload + migrations):
npm run dev
- Frontend-only dev:
npm run dev:react(uses Vite) - Backend webserver dev:
npm run dev:webserver(runssrc/backend/webserver/main.tsviatsx) - Build production bundles:
npm run build(runsbuild:react,build:preload,build:migrations,build:electron) - Package for Windows:
npm run package(callselectron-builder) - Tests:
npm test(runsvitest), coverage:npm run test:coverage
Important patterns and conventions
- Multi-config Vite builds: see
vite.electron.config.ts,vite.preload.config.ts, andvite.migrations.config.ts. When adding runtime files for main/preload/migrations, update those configs. - IPC surface is declared in
src/preload/preload.tsaswindow.electronAPI. When adding or renaming IPC channels:- Add/remove the
ipcMainhandlers in the Electron main process (src/backend/main) - Mirror the channel name and typings in
src/preload/preload.tsand update related renderer types undersrc/renderer/shared/types - Keep the exposed function names stable (renderer code expects methods like
electronAPI.getAllInvoices,electronAPI.addInvoice).
- Add/remove the
- Web vs Electron mode:
- Renderer detects runtime via
isWebMode()(seesrc/renderer/shared/api/restApi) and either calls the web API or talks towindow.electronAPI. - Mocking: MSW is enabled via
VITE_ENABLE_MOCKS(set in.env.*), and the renderer starts the worker insrc/renderer/main.tsx.
- Renderer detects runtime via
- Database & migrations:
- Migrations live in
src/backend/shared/migrationsand are built intodist-be/backend/migrations. Usenpm run build:migrationsfor CI packaging. - The app uses SQLite (
sqlite3). Be careful when changing schema—migrations need to run in a controlled order.
- Migrations live in
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.
- 4d ago First seen · 83 lines · 1,320 tokens per session scan A cda7680fa906
invoice-builder copilot-instructions.md is an instructions file published in the GitHub repository piratuks/invoice-builder (978 stars, last pushed today), licensed MIT. It adds 1,320 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.
Other instructions, from other repositories
sonicjs CLAUDE.md
Claude Code instructions for SonicJs-Org/sonicjs, covering sonicjs ai development guidelines, core technology stack, workspace boundary (conductor), architecture direction: document model (authoritative) and the 5 document tables (migration 0002documents.sql).
counterpoise-ledger CLAUDE.md
Instructions for jeffjjohnston/counterpoise-ledger, covering claude.md, project overview, development commands, essential commands and testing.
financial-analyst-skills AGENTS.md
Instructions for Ruinius/financial-analyst-skills, covering project documentation index, folder structure, skills overview and local server ports.
rodya-caijing-studio AGENTS.md
AGENTS.md instructions for nekopunch11/rodya-caijing-studio, covering rodya-caijing-studio · 给 ai agent 的使用说明, 入口与路由, 不可破的硬约束(合规灵魂,破了整套工具就失效), docx 渲染路由(财经内容台例外) and 在 codex 中安装.
mad-invoice-mcp CLAUDE.md
Claude Code instructions for mad-sol-dev/mad-invoice-mcp, covering claude.md, project overview, development commands, start the server and docker.
hogwarts CLAUDE.md
Claude Code instructions for databayt/hogwarts, covering claude.md, quick start, the 9 critical rules, database safety (critical) and never do these (auto-blocked by hooks).