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/lexmount/browseruse-agent-bench/copilot-instructionsgit clone --depth 1 https://github.com/lexmount/browseruse-agent-benchWhat 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 | $0.03057 | $0.03057 |
| Opus 5 | $0.01528 | $0.01528 |
| Sonnet 5 | $0.00611 | $0.00611 |
| Haiku 4.5 | $0.00306 | $0.00306 |
Grade A, and why
browseruse-agent-bench 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 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 — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generic Code Review Instructions
Comprehensive code review guidelines for GitHub Copilot that can be adapted to any project. These instructions follow best practices from prompt engineering and provide a structured approach to code quality, security, testing, and architecture review.
Review Language
When performing a code review, respond in Chinese.
Review Priorities
When performing a code review, prioritize issues in the following order:
🔴 CRITICAL (Block merge)
- Security: Vulnerabilities, exposed secrets, authentication/authorization issues
- Correctness: Logic errors, data corruption risks, race conditions
- Breaking Changes: API contract changes without versioning
- Data Loss: Risk of data loss or corruption
🟡 IMPORTANT (Requires discussion)
- Code Quality: Severe violations of SOLID principles, excessive duplication
- Test Coverage: Missing tests for critical paths or new functionality
- Performance: Obvious performance bottlenecks (N+1 queries, memory leaks)
- Architecture: Significant deviations from established patterns
🟢 SUGGESTION (Non-blocking improvements)
- Readability: Poor naming, complex logic that could be simplified
- Optimization: Performance improvements without functional impact
- Best Practices: Minor deviations from conventions
- Documentation: Missing or incomplete comments/documentation
General Review Principles
When performing a code review, follow these principles:
- Be specific: Reference exact lines, files, and provide concrete examples
- Provide context: Explain WHY something is an issue and the potential impact
- Suggest solutions: Show corrected code when applicable, not just what's wrong
- Be constructive: Focus on improving the code, not criticizing the author
- Recognize good practices: Acknowledge well-written code and smart solutions
- Be pragmatic: Not every suggestion needs immediate implementation
- Group related comments: Avoid multiple comments about the same topic
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 · 393 lines · 3,057 tokens per session scan A 42229c1d4adc
browseruse-agent-bench copilot-instructions.md is an instructions file published in the GitHub repository lexmount/browseruse-agent-bench (19 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 3,057 tokens to every session, about $0.0153 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
DM-Code-Agent AGENTS.md
Instructions for hwfengcs/DM-Code-Agent, covering agents.md, project intent, development rules, verification and important modules.
WorldCupArena CLAUDE.md
Instructions for wzk1015/WorldCupArena, covering project goal, key files, development process, coding style and running the website locally.
hermes-skilleval AGENTS.md
Instructions for Raidriar7170/hermes-skilleval, a project described as: Verification-gated skill routing and self-improvement harness for Hermes-style agent skills.
Browser4 CLAUDE.md
Instructions for platonai/Browser4, covering browser4 — project context for claude, architecture, key dispatch chain (cli → browser), batch commands and e2e test structure.
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.
wayland-core copilot-instructions.md
Instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.