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/tikoci/mikropkl/copilot-instructionsgit clone --depth 1 https://github.com/tikoci/mikropklWrote 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/instructions/tikoci/mikropkl/copilot-instructions)<a href="https://agentmods.dev/instructions/tikoci/mikropkl/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/tikoci/mikropkl/copilot-instructions.svg" alt="Measured on agentmods" 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 | $0.01296 | $0.01296 |
| Opus 5 | $0.00648 | $0.00648 |
| Sonnet 5 | $0.00259 | $0.00259 |
| Haiku 4.5 | $0.00130 | $0.00130 |
Grade A, and why
mikropkl 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 4d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Instructions — mikropkl
What This Project Does
mikropkl builds declarative virtual machine packages for MikroTik RouterOS CHR using pkl. The primary output is UTM bundles (.utm directories) for macOS, with QEMU launch scripts (qemu.sh + qemu.cfg) for Linux and CI testing.
Read CLAUDE.md first — it is the comprehensive project reference. AGENTS.md has deeper architecture decisions and investigation history for multi-session work.
Core Concepts
- Declarative configuration:
pklmanifests in/Manifestsdefine machines.makebuilds them to/Machines. TheMakefileis the source of truth for building. - Two boot tracks for CHR:
- SeaBIOS track (
*.qemu.*): Standard MikroTik x86 image with proprietary boot sector. QEMU loads via SeaBIOS chain-load. Simplest and fastest. - EFI/VirtIO track (
*.apple.*): Repackaged FAT16 EFI image (fromtikoci/fat-chrfor x86, standard for aarch64). Uses UEFI firmware (OVMF or EDK2). Required for Apple Virtualization.framework. The QEMU versions of*.apple.*bundles mirror the same pure-VirtIO scheme that Virtualization.framework uses.
- SeaBIOS track (
- UTM-first, QEMU second: UTM is the primary user experience on macOS. QEMU scripts make the same bundles work on Linux and in CI. A future CLI tool may manage QEMU-based instances from
~/.mikropkl/or~/.local/. - Platforms: macOS and Linux. Windows/WSL is not a focus — we won't document setup steps, but won't reject bug reports either.
Project Layout
| Directory | Purpose |
|---|---|
Manifests/ |
One .pkl per machine variant — what gets built |
Templates/ |
Mid-level pkl templates (amend Pkl/utmzip.pkl) |
Pkl/ |
Core pkl modules: utmzip.pkl, QemuCfg.pkl, CHR.pkl, UTM.pkl, etc. |
Machines/ |
Build output (git-ignored). One .utm directory per manifest |
Files/ |
Static files and user-facing supplementary docs (QEMU.md, efi_vars.fd) |
Lab/ |
Experiments, investigations, debug scripts — NOT production code |
.github/workflows/ |
CI: chr.yaml (build+release), auto.yaml (version detection), qemu-test.yaml (QEMU validation) |
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.
- 4d ago First seen · 62 lines · 1,296 tokens per session scan A b1385311df5c
mikropkl copilot-instructions.md is an instructions file published in the GitHub repository tikoci/mikropkl (52 stars, last pushed yesterday), licensed Unlicense. It adds 1,296 tokens to every session, about $0.0065 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
mikrotik-routeros-mcp AGENTS.md
AGENTS.md instructions for drohi-r/mikrotik-routeros-mcp, covering agents.md, project, core rules, key commands and key paths.
routeros-inspector-mcp CLAUDE.md
Claude Code instructions for kevinb361/routeros-inspector-mcp, covering routeros inspector mcp, purpose, hard safety boundary, public-repository hygiene and references.
routeros-inspector-mcp AGENTS.md
AGENTS.md instructions for kevinb361/routeros-inspector-mcp, a project described as: Unofficial read-only MCP server for MikroTik RouterOS inspection and policy-backed audits.
tik4net AGENTS.md
AGENTS.md instructions for danikf/tik4net, covering agents.md, what this is, reference map — read the right file, don't guess, build and tests.
mikrotik-mcp CLAUDE.md
Instructions for mikrotik-mcp/mikrotik-mcp, covering claude.md, what this is, commands, architecture and adding or changing a tool module.
br3eze-code CLAUDE.md
Instructions for br3eze-code/br3eze-code, covering claude.md, commands, what this project is, architecture — the live spine vs. legacy code and other real subsystems.