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/malwarebo/nyrve/interactivegit clone --depth 1 https://github.com/malwarebo/nyrveWhat 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 | $0.00938 | $0.00938 |
| Opus 5 | $0.00469 | $0.00469 |
| Sonnet 5 | $0.00188 | $0.00188 |
| Haiku 4.5 | $0.00094 | $0.00094 |
Grade A, and why
nyrve interactive.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 yesterday.
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.
This is a copy
100% identical to vscode interactive.instructions.md — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Interactive Window
The interactive window component enables extensions to offer REPL like experience to its users. VS Code provides the user interface and extensions provide the execution environment, code completions, execution results rendering and so on.
The interactive window consists of notebook editor at the top and regular monaco editor at the bottom of the viewport. Extensions can extend the interactive window by leveraging the notebook editor API and text editor/document APIs:
- Extensions register notebook controllers for the notebook document in the interactive window through
vscode.notebooks.createNotebookController. The notebook document has a special notebook view typeinteractive, which is contributed by the core instead of extensions. The registered notebook controller is responsible for execution. - Extensions register auto complete provider for the bottom text editor through
vscode.languages.registerCompletionItemProvider. The resource scheme for the text editor isinteractive-inputand the language used in the editor is determined by the notebook controller contributed by extensions.
Users can type in code in the text editor and after users pressing Shift+Enter, we will insert a new code cell into the notebook document with the content from the text editor. Then we will request execution for the newly inserted cell. The notebook controller will handle the execution just like it's in a normal notebook editor.
Interactive Window Registration
Registering a new editor type in the workbench consists of two steps:
- Register an editor input factory which is responsible for resolving resources with given
globpatterns. Here we register anInteractiveEditorInputfor all resources withvscode-interactivescheme. - Register an editor pane factory for the given editor input type. Here we register
InteractiveEditorfor our own editor inputInteractiveEditorInput.
The workbench editor service is not aware of how models are resolved in EditorInput, neither how EditorPanes are rendered. It only cares about the common states and events on EditorInput or EditorPane, i.e., display name, capabilities (editable), content change, dirty state change. It's EditorInput/EditorPane's responsibility to provide the right info and updates to the editor service. One major difference between Interactive Editor and other editor panes is Interactive Window is never dirty so users never see a dot on the editor title bar.
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.
- yesterday First seen · 51 lines · 938 tokens per session scan A 26ddeb782fd4
nyrve interactive.instructions.md is an instructions file published in the GitHub repository malwarebo/nyrve (5 stars, last pushed 2mo ago), licensed MIT. It adds 938 tokens to every session, about $0.0047 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to vscode interactive.instructions.md, differing in 0 lines, and is treated as a copy.
Other instructions, from other repositories
vibeyard CLAUDE.md
Instructions for elirantutia/vibeyard, covering claude.md, what this is, build & run, testing and architecture.
coder AGENTS.md
AGENTS.md instructions for coder/coder, covering coder development guidelines, task-specific guidance, workflow, essential commands and repository guardrails.
denova AGENTS.md
Instructions for alfredxw/denova, covering 项目目标, 必须遵守, 项目约定, 代码注意事项 and 前端与交互规范.
spring-plugin AGENTS.md
Instructions for explyt/spring-plugin, covering agents.md — contributor guide, project structure, build, test, run, code style and conventions and commits and pull requests.
coder CLAUDE.md
Claude Code instructions for coder/coder, a project described as: Secure environments for developers and their agents.
ORG2 AGENTS.md
AGENTS.md instructions for org2AI/ORG2, covering agents.md — agent skill routing for orgii, skill routing table, root-cause-first bug fixing, default delivery flow and ui copy conventions.