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/stilero/bankan/agents-mdgit clone --depth 1 https://github.com/stilero/bankanWrote 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/stilero/bankan/agents-md)<a href="https://agentmods.dev/instructions/stilero/bankan/agents-md"><img src="https://agentmods.dev/badge/instructions/stilero/bankan/agents-md.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 | $0.00850 | $0.00850 |
| Opus 5 | $0.00425 | $0.00425 |
| Sonnet 5 | $0.00170 | $0.00170 |
| Haiku 4.5 | $0.00085 | $0.00085 |
Grade A, and why
bankan AGENTS.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 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.
How it starts
The opening of the file, as written. The whole thing — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
This repository is split into two Node packages. client/ contains the Vite + React dashboard UI, with app code in client/src/ and production output in client/dist/. server/ contains the Express + WebSocket backend in server/src/. Root-level scripts/ holds setup utilities, docs/ is for supporting documentation, and .data/ stores local runtime state such as settings and workspaces. Keep UI components in PascalCase files such as KanbanBoard.jsx; keep backend modules small and role-based, such as agents.js or orchestrator.js.
Build, Test, and Development Commands
Run npm run setup for the interactive bootstrap flow; it writes .env.local and installs dependencies. Use npm start or npm run dev from the repo root to run server and client together. Use npm run dev --prefix client to work on the UI only, npm run build --prefix client to create a production bundle, and npm run dev --prefix server to run the backend with file watching. npm run install:all reinstalls root, server, and client dependencies.
Use npm run lint for the shared ESLint standard, npm run lint:fix for safe autofixes, npm test for the full automated suite, npm run test:server and npm run test:client for package-local runs, and npm run coverage to enforce the combined coverage gate used in CI.
Coding Style & Naming Conventions
Match the existing code style: ES modules, semicolons, single quotes, and concise functional React components. Use PascalCase for React component files, camelCase for hooks and helpers such as useFactory.js, and lowercase role-based names for backend modules. Follow the existing indentation and inline-style patterns already used in client/src/App.jsx and server/src/index.js. ESLint is the checked-in source of truth for shared code standards, so keep changes lint-clean.
Testing Guidelines
Follow TDD for new work and bug fixes: write or update the failing automated test first, confirm it fails for the expected reason, then implement the behavior and rerun the relevant package suite before finishing.
Agent-driven changes should follow the same order: tests first, implementation second, verification last. Do not start code changes before the target test exists unless the task is strictly documentation or configuration with no executable behavior.
Prioritize behavioral tests around workflow state, settings normalization, WebSocket state transitions, and critical UI interactions rather than asserting incidental implementation details.
Run npm run lint after code changes and before handing work off. Run npm test before handing work off. Run npm run coverage before opening a PR when your change affects covered logic. If coverage gaps remain intentional, document them in the PR.
For UI changes, keep manual verification through the dashboard at http://localhost:5173 for interaction polish, but automated tests are the default quality gate. If you add tests, place them next to the affected package and use clear names like feature-name.test.js.
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.
- 5d ago First seen · 28 lines · 850 tokens per session scan A 2b432fe16379
bankan AGENTS.md is an instructions file published in the GitHub repository stilero/bankan (4 stars, last pushed 5mo ago), licensed MIT. It adds 850 tokens to every session, about $0.0042 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.
Other instructions, from other repositories
Skill_Seekers CLAUDE.md
Instructions for yusufkaraaslan/Skill_Seekers, covering claude.md, project overview, essential commands, required before running tests or cli (src/ layout) and run all tests (never skip - all must pass before commits).
squarebox CLAUDE.md
Instructions for SquareWaveSystems/squarebox, covering claude.md, agent skills, issue tracker, triage labels and domain docs.
haive AGENTS.md
AGENTS.md instructions for martinambrus/haive, covering haive, stack, monorepo layout, package boundaries and architecture summary.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
Osverse AGENTS.md
AGENTS.md instructions for Oswald-Hao/Osverse, covering osverse repository rules for coding agents and protected-branch workflow.
substack-mcp-plus CLAUDE.md
Instructions for ty13r/substack-mcp-plus, covering claude.md, quick command reference (use these!), always activate virtual environment first, run tests (use python3!) and code quality.