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/microsoft/winappcli/agents-mdgit clone --depth 1 https://github.com/microsoft/winappCliWhat 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 | $0.03837 | $0.03837 |
| Opus 5 | $0.01919 | $0.01919 |
| Sonnet 5 | $0.00767 | $0.00767 |
| Haiku 4.5 | $0.00384 | $0.00384 |
Grade A, and why
winappCli 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 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 — 308 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent instructions for winapp
This file provides focused, actionable information to help an AI coding agent be immediately productive in this repo.
Audience contracts
User documentation
Write for someone using winapp. Lead with the smallest command or input that demonstrates the behavior, then answer: what can I do, what do I type, what happens, and what action should I take if it fails. Describe observable behavior, not implementation machinery.
Keep internal deliberation, review provenance, rejected alternatives, review rounds/IDs, resolved blockers, and defenses of the implementation out of user documentation. State each user-facing fact once on its canonical surface and link to it elsewhere instead of copying it.
Example: "After cloning a project with winapp.yaml, run winapp restore to
reinstall its pinned SDK packages and projections without changing versions. For
a .NET project without winapp.yaml, run dotnet restore instead."
Reviews and author communication
Assume the reader has no prior codebase or conversation context. A junior developer should understand each finding after one read. Use a plain title, then:
- What is wrong: the behavior or design defect.
- Show me: the smallest concrete command, input, tree, or code path, preferably as input -> actual result -> expected result.
- Why it matters: the user, build, or maintenance consequence.
- Smallest fix: the least-complex change that resolves it.
Define unavoidable jargon at first use. Put paths and lines in supporting detail, not in a title the reader must decode. If the issue cannot be demonstrated concretely, lower its confidence or drop it.
Example:
- What is wrong:
--setup-sdks nonefails instead of skipping SDK updates. - Show me:
winapp init --setup-sdks none-> exits nonzero in the package update loop; expected: succeeds without updating SDKs. - Why it matters: documented automation stops before initialization.
- Smallest fix: skip the package-update loop when the value is
none.
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 · 308 lines · 3,837 tokens per session scan A b7bcc2c9c5e9
winappCli AGENTS.md is an instructions file published in the GitHub repository microsoft/winappCli (1,205 stars, last pushed 4d ago), licensed MIT. It adds 3,837 tokens to every session, about $0.0192 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
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode tree-widgets.instructions.md
Use when asked to consume workbench tree widgets in VS Code.
vscode accessibility.instructions.md
Use when implementing accessibility features, ARIA labels, screen reader support, accessible help dialogs, or keybinding scoping for accessibility. Covers AccessibleContentProvider, CONTEXTACCESSIBILITYMODEENABLED, verbosity settings, and announcement patterns.
vscode writing-tests.instructions.md
VS Code test writing guidelines — unit tests, integration tests, snapshot tests, and clean teardown patterns. Reference when writing or updating tests.
vscode best-practices.instructions.md
VS Code best practices — reusing common UI primitives and patterns. Reference when writing or reviewing code.
vscode coding-guidelines.instructions.md
VS Code coding guidelines — naming, style, types, strings, and code quality rules. Reference when writing or reviewing code.