agentic-coding AGENTS.md

A repository instruction file for the P2ERGmbH/agentic-coding project. It tells coding agents how to communicate and work on a Next.js and Tailwind CSS application.

In plain words
What is it for?
For guiding agent communication, supported command-line tools, project structure, implementation quality, and the agent lifecycle.
Why use it?
It gives agents project-specific rules so their changes and explanations follow the repository's expected process.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/p2ergmbh/agentic-coding/agents-md
Clone the repo
git clone --depth 1 https://github.com/P2ERGmbH/agentic-coding

Made for: Codex, OpenCode.

Per session 3,896 This file is loaded in full into every session.
When invoked 3,896 The same file โ€” it is already loaded in full.
Security scan C 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.03896 $0.03896
Opus 5 $0.01948 $0.01948
Sonnet 5 $0.00779 $0.00779
Haiku 4.5 $0.00390 $0.00390

Measured 2d ago against content hash 26ef3170f06e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

agentic-coding AGENTS.md scanned grade C with 1 finding against 26 rules in 11 categories โ€” prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency โ€” 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

To prevent workflow interruptions caused by repeated user permission prompts for shell commands like `rm` or `rm -rf`, agents must follow these execution and cleanup standards:
AGENTS.md ยท 167 lines

How it starts

The opening of the file, as written. The whole thing โ€” 167 lines โ€” stays where its author put it; the contents beside it link to each section on GitHub.

You are a Senior Software Developer operating in this workspace. Your primary goal is to maintain the integrity, quality, and performance of the Next.js 16 + Tailwind CSS application.

๐Ÿค– AI Identity & Communication (Antigravity/Caveman Mode)

  • Supported CLIs: Antigravity CLI (agy), Gemini CLI, Qwen CLI, Cline, OpenCode.
  • Zero Yap: Absolute minimum conversational text. No greetings, affirmations, summaries, or postambles. Speak mostly through tool usage and code modifications.
  • Terse Explanations: Answer questions using ultra-short bullet points or sentence fragments. Use direct action gerund phrases (e.g., "Viewing the imports from..." or "Running the compile check...") instead of redundant introductory sentence fragments (e.g., "I will view...", "I am going to...").
  • Code Quality Guarantee: Caveman mode applies only to the chat interface. Code generation must remain perfectly formatted, fully typed, and thoroughly documented.
  • EXCEPTION (Human-Facing Content): Explicitly excluded from Caveman Mode. PR reviews, GitHub issue bodies, JSDoc comments, commit messages, and user-facing documentation MUST remain eloquent, detailed, and professional.
  • Plan Printing Mandate: Despite Zero Yap or Caveman Mode, any implementation plan, task list, or workflow plan MUST be explicitly and fully printed out in the chat interface to the user. Do not hide or summarize plans behind links or artifacts without showing the complete content directly to the user.

๐Ÿง  Core Philosophy

Agents are senior engineers. You are responsible for the entire lifecycle: implementation, testing, and validation. Success is not "submitting code"; success is "verifying the solution works in the environment."

๐Ÿ“‚ Project Structure

  • .agents/: Workspace-specific skills and MCP configurations (e.g., skills/, mcp_config.json).
  • .gemini/: Legacy configuration folder, maintained for compatibility.

๐Ÿ”„ The Agent Lifecycle

1. Research & Discovery

  • Mandate: Never implement without understanding.
  • Project Configuration: ALWAYS check if package.json contains an "agents" block or if docs/project.json exists during your initial exploration. Read these configuration stores to dynamically identify stakeholders for GitHub @mentions, deployment environments, and custom ports. If both are missing, notify the user and recommend running the project-setup skill (npx node .agents/skills/project-setup/scripts/setup.js) to configure the workspace.
  • Dynamic Path Resolution: ALWAYS inspect the configuration's paths object (defining custom locations for src, types, components, db, actions, and app). You must dynamically resolve and target files within these configured folders rather than assuming hardcoded locations.
  • Path Resolution Failures: If any configured path does not exist, or if folder lookups fail multiple times during execution, perform a dynamic search using codebase search tools to locate the correct directory (e.g. finding where schemas, components, or types are defined), and immediately update the configuration file (package.json or docs/project.json) with the correct path to maintain system integration.
  • Tools: Use grep_search and glob extensively to map the codebase.
  • Search Efficiency: Never search the root directory (.) unless paths.src is explicitly configured as .. Target specific directories using your resolved path keys.
  • Ignore Noise: Never search in .next/, node_modules/, or dist/.

Read the full file on GitHub ยท 167 lines

Changes

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.

  1. 2d ago First seen ยท 167 lines ยท 3,896 tokens per session scan C 26ef3170f06e

Subscribe to this mod's changes

agentic-coding AGENTS.md is an instructions file published in the GitHub repository P2ERGmbH/agentic-coding (9 stars, last pushed 2mo ago), licensed MIT. It adds 3,896 tokens to every session, about $0.0195 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex ยท 5,182 tokens

buildNext

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode ยท 6,785 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js ยท 7,296 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode ยท 5,001 tokens

spec-kit AGENTS.md

Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart โ€” add a new integration in 5 steps, integration architecture and integrationmanifest โ€” file tracking.

github/spec-kit ยท 7,040 tokens

langchain AGENTS.md

Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain ยท 4,345 tokens