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/sidrat2612/agent-policykit/copilot-instructionsgit clone --depth 1 https://github.com/sidrat2612/agent-policykitWhat 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 | $0.05039 | $0.05039 |
| Opus 5 | $0.02520 | $0.02520 |
| Sonnet 5 | $0.01008 | $0.01008 |
| Haiku 4.5 | $0.00504 | $0.00504 |
Grade A, and why
agent-policykit 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 — 524 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Source of Truth
The Miro board is the single source of truth for all product, design, and architecture decisions:
- Board URL: https://miro.com/app/board/uXjVHSJQ-Eg=/
- Never guess or assume project-specific details (features, specs, data, numbers, flows, designs)
- Check the extracted docs first in
docs/before answering questions about the product - If the required information is missing, ambiguous, or not covered in
docs/, then check the Miro board using MCP tools (context_explore,context_get,table_list_rows) - If the information is not present on the Miro board, ask the user — do not fabricate or approximate
Project Context
- Platform: Life-Path Mapping Platform for Indian students (14–24)
- Stack: Next.js (web), React Native + Expo (mobile), NestJS (backend), PostgreSQL, Redis
- Docs folder:
docs/contains extracted specs from Miro (numbered 01–21) - Treat student data and any under-18 user flows as high-scrutiny from a privacy, consent, retention, and access-control perspective
Theme Governance
- The finalized Professional Growth Palette is the source of truth for all visual theme work across the website, webapp frontend, and mobile frontend.
- Theme tokens must be maintained centrally. Do not define brand, track, semantic, or status colors inside individual pages or components.
- Website theme sources must remain centralized in the shared website theme files and token roots.
- Webapp frontend theme sources must remain centralized in the shared webapp theme files and token roots.
- Mobile frontend theme sources must follow the same rule: maintain one central theme source and consume it everywhere.
- If a new visual state or color is needed, add it to the central theme first, then consume that token in pages and components.
- Avoid hardcoded hex values and ad hoc Tailwind utility colors for brand or semantic styling when a centralized token should exist.
Engineering Governance Standard
These instructions govern code generation, refactoring, reviews, infrastructure automation, CI/CD, prompts, queries, architecture proposals, and other AI-assisted engineering work in this repository.
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 · 524 lines · 5,039 tokens per session scan A 7c601c97067c
agent-policykit copilot-instructions.md is an instructions file published in the GitHub repository sidrat2612/agent-policykit (5 stars, last pushed 3mo ago), licensed MIT. It adds 5,039 tokens to every session, about $0.0252 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
agent-skills-lint AGENTS.md
Instructions for swarmclawai/agent-skills-lint, covering agent-skills-lint — agent-facing reference, global flags, json envelope, exit codes and commands.
routesmith copilot-instructions.md
Instructions for sidrat2612/routesmith, covering routesmith - copilot instructions, core architecture rule, key design principles, versioning rule and description alignment rule.
coding-agent-eval CLAUDE.md
Claude Code instructions for ttxs69/coding-agent-eval, covering claude.md, what this is, environment, commands and run a single test.
helmor AGENTS.md
Instructions for dohooo/helmor, covering agents.md, what is helmor, commands, architecture and three-process model.
matt-skills-with-to-goal CLAUDE.md
Instructions for tt-a1i/matt-skills-with-to-goal: Skills are organized into bucket folders under skills/.
ephemeral-sandbox CLAUDE.md
Instructions for Ephemeral-AI-Lab/ephemeral-sandbox, covering claude.md, project, engineering practice (required), build & test and sandbox tools.