MicrosoftKnown publishermicrosoft/agentic-journeys/journeys/smart-todo/setup/copilot-instructions.md

agentic-journeys / setup

Copilot instructions5 starsChanged 46 days agoMITView on GitHub

Instructions for every Copilot session in this repository. Phase 4 adds a "Definition of Done" section. The hook in .github/hooks/tdd-guard.json blocks changes to these files:

Where files are headingNew Copilot instructions files have more rules about branches and pull requests, and fewer about code style.
Of the Copilot instructions files we read, the 2,336 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- container-apps-deploymentSkill
- data-access-abstractionSkill
- grafana-azureSkill
- grill-planSkill
- journey-runnerSkill
- journey-templateSkill
- journey-test-harnessSkill
- n8n-azureSkill
- superset-azureSkill
- flex-functions-sql-foundrySkill

The file

We split this copilot-instructions.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific
could be automatic: a setting could enforce it

Very specific:
Use the grill-plan skill before code exists, and the tdd-builder agent for the red and green phases.
line 16
Very specific:
Never commit or print secrets, local.settings.json, or .azure/.
line 30

≈ “Common” is an estimate: at least 1 in 5 of the 3,779 Copilot instructions files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

copilot-instructions.md

SmartTodo

Instructions for every Copilot session in this repository. Phase 4 adds a "Definition of Done" section.

Map

journeys/smart-todo/PLAN.md: the vision, the quality gates, CI, review triage, and the stack of pull requests. The phase plans are PLAN-phase1-api.md to PLAN-phase4-factory.md in the same folder.
journeys/smart-todo/src/api: the Azure Functions API (Node.js, TypeScript, and Vitest).
journeys/smart-todo/src/ios: the SwiftUI app, which starts as a copy of starter/ios.
journeys/smart-todo/infra, azure.yaml, and scripts/: the infrastructure and its gates.
.github/scripts/verify-smart-todo.mjs: the checked-in black-box verifier.

Workflow

The plans are the source of truth. Update the plan section before you change behavior.
Use the grill-plan skill before code exists, and the tdd-builder agent for the red and green phases.
Red: commit only failing tests and the stubs they need. Green: make the tests pass without changing them.
Work is done only when the commands in the "Quality Gates" section of journeys/smart-todo/PLAN.md exit 0. Run each command from the directory that the section names.
Phases 1 to 3 are one stack of pull requests that gh stack manages in the current checkout. Don't put stack layers in separate worktrees. Don't merge pull requests; the human merges them.

Protected files

The hook in .github/hooks/tdd-guard.json blocks changes to these files:

Tests that a red tag (phase1-red, phase2-red, or phase3-red) freezes. If a frozen test is wrong, stop and report it. Never move or delete a red tag; the human does that.
.github/hooks/, .github/scripts/verify-smart-todo.mjs, journeys/smart-todo/scripts/test-ios.mjs, and journeys/smart-todo/starter/.

Always

Never commit or print secrets, local.settings.json, or .azure/.
Commands must work in PowerShell, Command Prompt, bash, and zsh. Don't use shell command substitution.
List problems that you hit and fixed in the pull request description, under "Problems and fixes".

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free