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/speedpy/speedpy/agents-mdgit clone --depth 1 https://github.com/speedpy/speedpyWhat 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 | $0.20341 | $0.20341 |
| Opus 5 | $0.10171 | $0.10171 |
| Sonnet 5 | $0.04068 | $0.04068 |
| Haiku 4.5 | $0.02034 | $0.02034 |
Grade A, and why
speedpy 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sS -o speedpycom/data/disposable_email_blocklist.conf \ How it starts
The opening of the file, as written. The whole thing — 1,747 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This file is the source of truth for AI Coding Agents working in this repository.
CLAUDE.md and .cursorrules exist only as pointers to this file — keep guidance here.
Additional case-specific instructions
Read the matching file before you start, if your task is one of these. They live outside this file so that a capability most projects never touch does not cost every reader the tokens to scroll past it.
- video manipulation, transcoding, compression, video upload, ffmpeg:
agents_docs/working_with_video_files.md
Adding one: keep the file in agents_docs/, name it for the task rather than the
technology (working_with_<thing>.md), and add exactly one bullet here —
searchable words first, path last. Anything a project would only sometimes need
belongs there, not in this file.
Project Overview
SpeedPy Standard is a Django-based web application starter template featuring a single-app architecture with custom user authentication, Celery for background tasks, and Tailwind CSS for styling. The project supports two development modes — Docker Compose or local uv + npm — picked at init time.
Development Commands
For mode-specific run commands (Django management, Tailwind, migrations, tests, shell access, etc.) read
AGENTS-local.md in this directory. It is populated by init-docker.sh or init-local.sh and reflects the
active setup. Do not assume a particular wrapper (docker compose run vs uv run) — the cheat sheet is the source
of truth for invocation.
Initialization scripts:
bash init-docker.sh— boots the project with Docker Compose (Postgres, Redis, Celery, nginx media).bash init-local.sh— runs the project on the host with uv + npm, SQLite, no Redis, Celery in always-eager mode.
Architecture
Apps Structure (single-app architecture)
This is a single-app Django project. All business logic ships from mainapp.
Do not create a new Django app for a new feature, page, or model — extend mainapp.
The only existing apps and their narrow purposes:
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 · 1,747 lines · 20,341 tokens per session scan A eeb46ed1129e
speedpy AGENTS.md is an instructions file published in the GitHub repository speedpy/speedpy (79 stars, last pushed 2d ago), licensed MIT. It adds 20,341 tokens to every session, about $0.1017 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
django-starter-template AGENTS.md
Instructions for wilfredinni/django-starter-template, covering agents.md, commands, lint & typecheck, architecture and testing.
django-saas-boilerplate CLAUDE.md
Instructions for eriktaveras/django-saas-boilerplate, covering django saas boilerplate, tech stack, project structure, key conventions and user model.
django-project-template AGENTS.md
Instructions for rnegron/django-project-template, covering agents.md, project overview, essential commands, development setup and local development (without docker).
CoachSync AGENTS.md
Instructions for rakibul-islam-raju/CoachSync, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
django-project-template CLAUDE.md
Instructions for rnegron/django-project-template: This project uses AGENTS.md as the single source of truth for AI coding agent guidance. See that file for project overview, commands, architecture, and conventions.
mostlylucidweb CLAUDE.md
Claude Code instructions for scottgal/mostlylucidweb, covering claude.md, working preferences, project overview, build & development commands and .net backend.