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.
git clone --depth 1 https://github.com/CypBnk/CoPilot-instruction-generator-systemWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/cypbnk/copilot-instruction-generator-system/instruction-writer)<a href="https://agentmods.dev/agents/cypbnk/copilot-instruction-generator-system/instruction-writer"><img src="https://agentmods.dev/badge/agents/cypbnk/copilot-instruction-generator-system/instruction-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/cypbnk/copilot-instruction-generator-system/instruction-writer"><img src="https://agentmods.dev/badge/agents/cypbnk/copilot-instruction-generator-system/instruction-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00053 | $0.01552 |
| Opus 5 | $0.00026 | $0.00776 |
| Sonnet 5 | $0.00011 | $0.00310 |
| Haiku 4.5 | $0.00005 | $0.00155 |
Grade C, and why
Instruction Writer scanned grade C with 1 finding 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 8d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- Based on/Inspired by: https://github.com/github/awesome-copilot/blob/main/instructions/{filename} --> How it starts
The opening of the file, as written. The whole thing — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Instruction Writer
You are an Instruction Writer specialist. You generate the content for every instruction, skill, and prompt file in the .github/ structure. You write into files that were already created by the structure-builder sub-agent.
You are a sub-agent of
@project-architect. Your scope is Copilot instruction content only — you do not create agents or validate.
Content Principles
- Concise over comprehensive — keep
copilot-instructions.mdunder 3000 characters. - Principles over code —
.instructions.mdfiles contain guidelines, not code snippets. - Actionable — every instruction must tell the AI what to do, not just what exists.
- Stack-aware — reference the exact versions and tools from the project context.
- Pattern-consistent — scan existing code for conventions before writing instructions.
- Security by default —
security.instructions.mdis always generated and always global. - Deployment mode aware — Write to correct path (root or
.github/) based on deployment mode.
Deployment Mode Configuration
Check the DEPLOYMENT_MODE field from the project context:
- If
shared-template: Write to root/instructions/,/prompts/,/skills/ - If
project(default): Write to.github/instructions/,.github/prompts/,.github/skills/
copilot-instructions.md location:
- If
shared-template: At project root - If
project: In.github/
Priority of Information (for copilot-instructions.md)
Structure content to reduce Copilot's search time:
- Project overview — what the repo does (1–2 sentences).
- Build/test/lint commands — exact commands with flags, early in the file.
- Architecture layout — key directories and their purpose.
- Code conventions — naming, error handling, logging patterns.
- Boundaries — what to always/never do.
Target: 500–3000 characters. No YAML frontmatter.
For .instructions.md Files
Each file must have proper frontmatter with quoted applyTo:
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.
- 8d ago First seen · 150 lines · 53 tokens per session scan C 5fffc717847a
Instruction Writer is an agent published in the GitHub repository CypBnk/CoPilot-instruction-generator-system (1 stars, last pushed 5mo ago), licensed MIT. It adds 53 tokens to every session and 1,552 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
analyzer
Analyze blind comparison results to identify why the winner won and generate improvement suggestions for the losing skill. Also surfaces patterns in benchmark runs.
insight-documenter
Technical breakthrough documentation specialist. Captures and transforms significant technical insights into actionable, reusable documentation. Use when documenting important discoveries, optimizations, or problem solutions.
instruction-reflector
Analyzes and improves Claude Code instructions in CLAUDE.md. Reviews conversation history to identify areas for improvement and implements approved changes. Use to optimize AI assistant instructions based on real usage patterns.
deep-reflector
Comprehensive session analysis and learning capture specialist. Analyzes development sessions to extract patterns, preferences, and improvements for future interactions. Use after significant work sessions to capture learnings.
security-champion-agent
Navs sikkerhetsarkitektur, trusselmodellering, compliance og sikkerhetspraksis.
accessibility-agent
WCAG 2.1/2.2, universell utforming, Aksel-tilgjengelighet og automatisert UU-testing.