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/arvo-ai/aurora/agents-mdgit clone --depth 1 https://github.com/Arvo-AI/auroraWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/arvo-ai/aurora/agents-md)<a href="https://agentmods.dev/instructions/arvo-ai/aurora/agents-md"><img src="https://agentmods.dev/badge/instructions/arvo-ai/aurora/agents-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.02731 | $0.02731 |
| Opus 5 | $0.01366 | $0.01366 |
| Sonnet 5 | $0.00546 | $0.00546 |
| Haiku 4.5 | $0.00273 | $0.00273 |
Grade A, and why
aurora 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 4d 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.
- Subprocess calls (`subprocess.run`, `Popen`, etc.) must receive an explicit `env` parameter with only the variables they need — never inherit the full server environment. How it starts
The opening of the file, as written. The whole thing — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Aurora - Agent Guidelines
Commands
- Start dev:
make dev(builds & starts all containers) - Stop:
make down - View logs:
make logs(shows last 50 lines, follows) - Rebuild API:
make rebuild-server(rebuild aurora-server only) - Frontend lint:
cd client && npm run lint - Frontend build:
cd client && npm run build - Backend logs:
docker logs -f aurora-celery_worker-1(orkubectl logs -f deployment/aurora-celery-worker) - Deploy with Docker Compose:
make prod(production build)
Docker deployments
- Development: Use
make dev-buildto build the project,make devto start containers, andmake downto stop them. - Production (prebuilt images): Use
make prod-prebuiltto pull from GHCR, retag, and run. Usemake prod-localto build from source instead. Usemake downto stop. - Production (build from source): Use
make prod-localormake prod-buildfor feature branch demos and custom builds. - Important: Always update both
docker-compose.yamlanddocker-compose.prod-local.ymltogether to keep environment variables in sync.
Architecture
- Docker Compose stack: aurora-server (Flask API on :5080), celery_worker (background tasks), chatbot (WebSocket on :5006), frontend (Next.js on :3000), postgres (:5432), weaviate (vector DB :8080), redis (:6379), vault (secrets :8200), seaweedfs (object storage :8333)
- Backend (server/): Flask REST API (main_compute.py), WebSocket chatbot (main_chatbot.py), Celery tasks, connectors for GCP/AWS/Azure/Datadog/New Relic/Grafana, LangGraph agent workflow
- Frontend (client/): Next.js 15, TypeScript, Tailwind CSS, shadcn/ui components, Auth.js authentication, path alias
@/*→./src/* - Database: PostgreSQL (aurora_db), Weaviate for semantic search, Redis for Celery queue
- Secrets: HashiCorp Vault (KV v2 engine at
auroramount) - Object Storage: S3-compatible via SeaweedFS (default), supports AWS S3, Cloudflare R2, MinIO, etc.
- Config: Environment in
./.env, GCP service account inserver/connectors/gcp_connector/*.json
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.
- 4d ago First seen · 154 lines · 2,731 tokens per session scan A 877d4f2ac892
aurora AGENTS.md is an instructions file published in the GitHub repository Arvo-AI/aurora (404 stars, last pushed yesterday), licensed Apache-2.0. It adds 2,731 tokens to every session, about $0.0137 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
awesome-kubernetes CLAUDE.md
Instructions for nubenetes/awesome-kubernetes, covering claude.md — nubenetes project instructions for claude code, git workflow: gitflow, branches, release process (mandatory for all releases) and versioning.
awesome-kubernetes GEMINI.md
Instructions for nubenetes/awesome-kubernetes, covering 🧠 core mandates, 🛠️ structural evolution & navigation, 📊 mermaid diagram best practices, 🛡️ repository policies & branch protection and 📝 readme synchronization & maintenance protocols.
IsaacAutomator AGENTS.md
AGENTS.md instructions for isaac-sim/IsaacAutomator, covering isaac automator - agent instructions, operator quickstart (the short version), rules for agents (read before acting) and map.
nudgebee CLAUDE.md
Claude Code instructions for nudgebee/nudgebee, covering nudgebee monorepo - development guide, module structure, ai coding principles, 2. definition of done and 3. fix the class, not the instance (rule of three).
nudgebee copilot-instructions.md
Copilot instructions for nudgebee/nudgebee, covering copilot instructions for llm server, project architecture, developer workflows, key conventions & patterns and integration points.
nudgebee AGENTS.md
AGENTS.md instructions for nudgebee/nudgebee: This file exists so AI coding assistants that look for AGENTS.md first (Aider, Cline, OpenCode, Codex CLI configs, and others adopting the cross-tool convention) find a project entrypoint without us maintaining a parallel document.