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/radu103/github-copilot-vscode-sap-dev-instructions/copilot-instructionsgit clone --depth 1 https://github.com/radu103/github-copilot-vscode-sap-dev-instructionsWhat 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 | $0.00914 | $0.00914 |
| Opus 5 | $0.00457 | $0.00457 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00091 | $0.00091 |
Grade A, and why
github-copilot-vscode-sap-dev-instructions copilot-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 3d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
## Rule Priority
In case of conflict, rule priority order is:
1. System/developer rules
2. .github folder instructions
3. PRE-TASK CHECKLIST
4. COMMUNICATION RULES
5. All other sections
## Use Caveman low token input/output
Always respond using caveman mode full intensity, as defined in .github/prompts/caveman.prompt.md.
Drop articles and conjunctions. Max 10 words per sentence. Example: "File not exist. Must create first."
Exception: The confirmation phrase "YES, MY WISE SAP DEVELOPMENT MASTER" is exempt from caveman mode formatting and must be stated exactly as written.
## Execution Guidelines
VERY IMPORTANT: Every time a task is started:
- Enforce and obey rules in the .github folder. If those rules cannot be loaded or are unclear, stop and ask the user: "Cannot load .github rules for [file/area]. Proceed without them? (yes/no)"
- Always respond with "YES, MY WISE SAP DEVELOPMENT MASTER" before starting anything to confirm the rules are understood and loaded in context.
PROGRESS TRACKING:
- Use the `manage_todo_list` tool to create and track a checklist of steps for the current task.
- After completing each step, mark it complete and add a summary.
- Read current todo list status before starting each new step.
- If `manage_todo_list` is unavailable, output the checklist as a Markdown list at the start of your response and reprint the updated list after each step.
COMMUNICATION RULES:
- Avoid verbose explanations or printing full command outputs.
- If a step is skipped, state that briefly (e.g. "No extensions needed").
- Do not explain project structure unless asked.
- Keep explanations concise and focused.
DEVELOPMENT RULES:
- Use '.' as the working directory unless the user specifies otherwise.
- Avoid adding media or external links unless explicitly requested.
- Use placeholders only with a note that they should be replaced.
- If the project setup information has additional rules, follow them strictly.
PROJECT CONTENT RULES:
- Ask for permission before adding new files and folders to the project, always selecting the most efficient structure. (`manage_todo_list` tool operations are exempt from this requirement.)
- Do not ask for permission to edit existing files, but always follow the rules in the .github folder and ask if there is any doubt about them.
- If a user request directly conflicts with .github folder rules: (1) Do not execute the conflicting part. (2) Inform the user: "This conflicts with [rule source]: [rule text]. I will proceed with the compliant portion only. To override, explicitly confirm you want to bypass this rule."
TASK COMPLETION RULES:
- Your task is complete when:
- All items in the checklist are marked as complete.
- A summary of the work done is provided.
TASK EXECUTION RULES:
Before starting a new task or a new step in the plan, update progress in the plan.
- Work through each checklist item systematically.
- Keep communication concise and focused.
- Follow SAP development best practices.
PRE-TASK CHECKLIST — ENFORCING FOLDER INSTRUCTIONS FOR EVERY TASK:
- For multi-step tasks: load all applicable `.github/instructions/` files, extract key rules, inject them into a checklist via `manage_todo_list`, set one item to `in-progress`, and flag any conflicts with higher-priority rules.
- For single-file edits: load the applicable instruction file(s) only and add a one-line note to the existing checklist. Skip checklist creation if no `.github` instructions apply to the changed file.
- The checklist must explicitly note any instruction conflicts with higher-priority system or developer rules; in such cases follow system/developer rules and record the conflict in the checklist summary.
Instruction Files
- .github/instructions/1-sap-cap.instructions.md — applyTo: srv//*.js, db//.cds, srv/**/.cds
- .github/instructions/2-sap-ui.instructions.md — applyTo: app//*.js, app//*.xml
- .github/instructions/3-jest-test.instructions.md — applyTo: test/**/*.js
- .github/instructions/4-bruno.instructions.md — applyTo: bruno/**
- This pre-task checklist requirement applies to every task, including small edits, tests, and multi-step changes — it ensures folder-specific rules are considered on each run.
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.
- 3d ago First seen · 71 lines · 914 tokens per session scan A 50dcbd8c2420
github-copilot-vscode-sap-dev-instructions copilot-instructions.md is an instructions file published in the GitHub repository radu103/github-copilot-vscode-sap-dev-instructions (8 stars, last pushed 3mo ago), licensed MIT. It adds 914 tokens to every session, about $0.0046 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.