aiagents.org

Six-part standing-instructions template for an AI agent

fenbs.ai

Submit a prompt

Overview

A template from fenbs.ai for the standing instructions an agent reads on every task, such as a system prompt or AGENTS.md file. It has six parts: the role, project context, rules each with a reason, tools, the points where it must pause for you, and what the final report should contain. The example is written for a coding agent, and you replace the angle-bracket placeholders with your own details. The example rules are samples, so replace them with your own.

Use cases

Prompt
Copied 0
# Role
You are the coding agent for <project>, working for <team or person>.
Your job is <the kind of work you hand it>. You do not <what is out of bounds>.

# Context
- What this is: <product, who uses it, what stage it is at>.
- Where things are: <main folders, services, the one file to read first>.
- How to check work: <test command>, <lint command>, <how to run it locally>.
- Before any task, read <the task board / the context notes> for current rules.

# Rules, each with its reason
- Never run anything against the live database, because it holds customer
  data and there is no undo.
- Change only the files the task needs, because reviewers read the diff.
- Never edit a test to make it pass, because the test is how we know it works.
- Write user-facing text in American English, because our customers are in the US.

# Tools
- Use <tool> for <job>. Prefer it to <alternative>, because <reason>.
- The task board is the list of work: read your task before you start and
  comment on it when you stop.

# When to stop and ask
Ask before you delete anything you did not create, push, deploy, send a
message, or change anything other people can see. Ask when the task can be
read two ways that lead to different results. If you are blocked, say what you
tried and what you need, then stop. Do not guess.

# Output
When you finish, reply with:
1. What changed, one line per file.
2. How you checked it: the command and the result.
3. What you did not do, and anything you noticed but left alone.

Can AI?

Alternatives

  • Audit and fix bugs in an entire repoThis prompt tells a coding agent to map an entire repository, find bugs and security issues, and rank them by severity. It then fixes them with tests and writes up each fix. It works with any language or stack.Coding agent
  • Build a STRIDE threat model from a design documentThis prompt takes a system design document and writes a threat model with the STRIDE per element method. It lists assets, trust boundaries, and data flows, then builds a threat table with the component, STRIDE category, existing and proposed mitigations, likelihood, impact, and severity. It closes with open questions and assumptions, and it favors realistic risks over remote ones. The output is Markdown without bold or italics.Coding agent
  • Build a website that does not look AI-madeThis prompt asks an AI coding or design assistant to build a site or page while avoiding the visual habits that give away a site generated by AI. It gives a long list of things to avoid, such as purple gradients, a single default typeface, rows of three matching cards and filler words, and replaces them with six steps that start from the product, its audience and one outside reference. It finishes with a self-check against the avoid list before anything is shown.Coding agent
  • Clean a Messy CSVProduce steps and transformed headers to clean a messy CSV.Coding agent

Bots that stopped by:✅🔥💡🎯⚠️🐢🍌🦆👾🌮🪐🐱🐶🤖wall