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/originalankur/generateagents.md/agents-mdgit clone --depth 1 https://github.com/originalankur/GenerateAgents.mdWhat 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 | $0.03246 | $0.03246 |
| Opus 5 | $0.01623 | $0.01623 |
| Sonnet 5 | $0.00649 | $0.00649 |
| Haiku 4.5 | $0.00325 | $0.00325 |
Grade A, and why
GenerateAgents.md AGENTS.md scanned grade A 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 3d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
* **Error Handling**: Application logic within modules and utilities should raise specific exceptions (e.g., `FileNotFoundError`, `subprocess.CalledProcessError`). Generic `except Exception` blocks should be avoided. A s How it starts
The opening of the file, as written. The whole thing — 278 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — AutogenerateAgentsMD.md
Project Overview
GenerateAgents.md is a Python command-line tool that automates the creation of a comprehensive AGENTS.md file for any public GitHub or local code repository. It acts as an automated codebase analyst and technical writer, using the dspy framework to programmatically interface with LLMs. The tool clones and analyzes a target codebase to produce a standardized blueprint, enabling AI coding agents to rapidly understand a project's architecture, conventions, and data flow. The primary language is Python (>=3.12).
Tech Stack
- Primary Language: Python (>=3.12)
- Core AI Framework:
dspy - LLM Abstraction Layer:
litellm - Dependency Management:
uv - CLI Framework:
argparse(standard library) - Configuration:
python-dotenv - Version Control Interaction:
git(viasubprocess) - Testing:
pytest
Architecture
The application follows a modular, stateless pipeline pattern orchestrated by the main CLI entry point.
src/autogenerateagentsmd/cli.py: The command-line interface entry point. It parses arguments and orchestrates the entire analysis and generation pipeline via therun_agents_md_pipelinefunction.src/autogenerateagentsmd/modules.py: Contains the coredspy.Moduleclasses (CodebaseConventionExtractor,AgentsMdCreator,AntiPatternExtractor). These modules encapsulate the primary LLM-driven logic for analyzing code and synthesizing the final document.src/autogenerateagentsmd/signatures.py: Defines the contracts for LLM interactions usingdspy.Signature. These signatures specify the expected inputs (e.g., source code) and outputs (e.g., extracted conventions) for each LLM-powered step.src/autogenerateagentsmd/model_config.py: Centralizes the configuration for supported LLMs, making it easy to switch between models like Gemini, Claude, and OpenAI.src/autogenerateagentsmd/utils.py: Contains helper functions for non-LLM tasks, such as cloning Git repositories, loading files into memory, and other file system operations.tests/: The test suite, containing end-to-end and unit tests.pyproject.toml: Defines project metadata, dependencies, and theautogenerateagentsmdconsole script entry point.
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.
- 3d ago First seen · 278 lines · 3,246 tokens per session scan A ec89263f5883
GenerateAgents.md AGENTS.md is an instructions file published in the GitHub repository originalankur/GenerateAgents.md (255 stars, last pushed 6mo ago), licensed MIT. It adds 3,246 tokens to every session, about $0.0162 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
claude-code-karma CLAUDE.md
Instructions for JayantDevkar/claude-code-karma, covering claude.md, project overview, quick start, clone (no submodule init needed — this is a monorepo) and start api (terminal 1).
agentic-workflow CLAUDE.md
Instructions for gtrabanco/agentic-workflow, covering claude.md, repository layout, working rules, authoring a skill and hand off, don't compose across a model/effort boundary.
gangsta AGENTS.md
Instructions for kucherenko/gangsta, covering contributing to gangsta, what this is, what will not be accepted, skill changes and the laws apply to contributors too.
gangsta GEMINI.md
Instructions for kucherenko/gangsta, a project described as: AI agentic skills framework for spec-driven development, built on the organizational model of mafia.
n-n-code-skills AGENTS.md
Instructions for n-n-code/n-n-code-skills, covering agents.md, repo purpose, structure, skill taxonomy and working rules.
agent-skills AGENTS.md
Instructions for rolandbrecht/agent-skills, covering agent instructions, 1. architectural changes & refactoring, 2. debugging & test failures, 3. structural code refactoring and 4. post-task reflection.