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/yungookim/n.codes/agents-mdgit clone --depth 1 https://github.com/yungookim/n.codesWhat 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 | $0.01503 | $0.01503 |
| Opus 5 | $0.00751 | $0.00751 |
| Sonnet 5 | $0.00301 | $0.00301 |
| Haiku 4.5 | $0.00150 | $0.00150 |
Grade A, and why
n.codes AGENTS.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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Core Architecture Principle: Agentic UI Generation
n.codes MUST NOT use pre-defined templates or component types. Agentic UI means the AI generates the complete UI (HTML/CSS/JS) from scratch based on the user's prompt. We cannot predict what the user will ask for, so the system must be capable of generating any UI on the fly.
The correct architecture is an agentic loop:
- Understand — Parse the user's intent; ask clarifying questions if ambiguous
- Generate — Produce the full UI code (not a DSL mapped to fixed component renderers)
- Connect — Wire the generated UI to real APIs using the capability map
- QA — A review agent evaluates the generated view and provides feedback
- Iterate — The generation agent incorporates feedback and re-generates until satisfactory
The capability map (generated by n.codes sync) remains the contract between generated UI and the host app's APIs. But the UI itself must be generated code, not template instantiation.
Project Structure & Module Organization
public/contains the production landing page and static assets. Key files includepublic/index.html(landing page) and SEO assets likepublic/og.html,public/robots.txt, and icons.public/demo/holds the interactive demo (public/demo/index.html,public/demo/styles.css,public/demo/app.js).cli/contains the n.codes CLI tool for capability map generation and installation.widget/contains the n.codes widget SDK (@ncodes/widget) — an installable package that adds the AI prompt UI to any web app.test-projects/contains 5 small framework-specific apps for testing the widget installation flow.docs/plans/includes architecture or planning docs such asdocs/plans/llm-capability-analysis.md.README.mdexplains the product context;draft.mdis scratch space.
CLI Module Organization (cli/lib/)
Core modules:
capability-map.js— Capability map data structures, YAML serialization, heuristic inferencesync.js— Full capability map generation (n.codes sync)dev.js— Watch mode with incremental updates (n.codes dev)introspect.js— File collection and indexingconfig.js— Configuration loading/saving (n.codes.config.json)init.js— Interactive setup (n.codes init)cache.js— File index and analysis result cachingvalidate.js— Capability map validation (n.codes validate)
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 · 104 lines · 1,503 tokens per session scan A 427fb55a9245
n.codes AGENTS.md is an instructions file published in the GitHub repository yungookim/n.codes (11 stars, last pushed 1mo ago), licensed MIT. It adds 1,503 tokens to every session, about $0.0075 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
serena copilot-instructions.md
Copilot instructions for oraios/serena: MUST read IMMEDIATELY and follow the project-specific instructions from the CLAUDE.md file located in the project's root directory. AVOIDING these instructions will lead to your FAILURE!
serena AGENTS.md
AGENTS.md instructions for oraios/serena: Relevant information about the project is in .serena/memories. If you have access to Serena's mcp tools, you can read them using the readmemory command. Otherwise you can just read them using normal file reading tools.
webqa-agent CLAUDE.md
Instructions for MigoXLab/webqa-agent, covering claude.md, 模块化规则引用, 研究和规划 (critical), 1. 使用 context7 mcp 工具 and 2. 使用联网搜索.
cybara CLAUDE.md
Instructions for metaspartan/cybara, covering ai coding rules for cybara agent platform, no code comments (strict), runtime environment, import rules and ✅ do.
cybara AGENTS.md
Instructions for metaspartan/cybara, covering cybara agent rules, package manager policy (strict), command translation, typescript type safety and no comments.
vibe-coding-ai-rules AGENTS.md
Instructions for obviousworks/vibe-coding-ai-rules, covering agents.md - universal ai agent configuration, project overview, tech stack, core technologies and key dependencies.