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/techequitycloud/rad-modules/agents-mdgit clone --depth 1 https://github.com/techequitycloud/rad-modulesWhat 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 | $0.13860 | $0.13860 |
| Opus 5 | $0.06930 | $0.06930 |
| Sonnet 5 | $0.02772 | $0.02772 |
| Haiku 4.5 | $0.01386 | $0.01386 |
Grade C, and why
rad-modules AGENTS.md scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
- Application Default Credentials are a single file shared machine-wide (`~/.config/gcloud/application_default_credentials.json`) — running `gcloud auth application-default login` in *any* terminal, including a different Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Diagnose: `curl -s "https://oauth2.googleapis.com/tokeninfo?access_token=$(gcloud auth application-default print-access-token)"` and compare the `email` field against who you expect to be authenticated as. How it starts
The opening of the file, as written. The whole thing — 559 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Workflows
This file contains workflow prompts for engineers to guide the agent. These workflows context-switch the agent into the specific mode required for different parts of the repository.
Global Workflow
Trigger: /global
Prompt:
You are an expert Senior DevOps Engineer specializing in Google Cloud Platform, GKE, and OpenTofu/Terraform. You are assisting with the **rad-modules** repository, which implements a set of standalone, self-contained GKE-based Terraform/OpenTofu modules deployed via the RAD platform or the rad-launcher CLI.
**Repository Structure:**
The repository is organized around eight independent modules under `modules/`. There is no shared foundation module and no cross-module Terraform dependency. Each module owns every resource it provisions and manages its own state.
| Module | What it provisions | Target audience |
|---|---|---|
| `Istio_GKE` | GKE Standard cluster + open-source Istio (sidecar or ambient mode) + Prometheus/Jaeger/Grafana/Kiali | Platform engineers learning upstream Istio |
| `Bank_GKE` | GKE cluster (Autopilot or Standard) + Cloud Service Mesh (managed Istio) + Bank of Anthos + optional Anthos Config Management + Cloud Monitoring SLOs | Engineers exploring ASM on a single cluster |
| `MC_Bank_GKE` | Multiple GKE clusters across regions + fleet-wide Cloud Service Mesh + Multi-Cluster Ingress + Multi-Cluster Services + Bank of Anthos behind a global HTTPS LB | Engineers exploring multi-cluster mesh and traffic |
| `AKS_GKE` | Azure AKS cluster registered with GCP as a GKE Attached Cluster via Fleet + GKE Connect agent via Helm | Engineers exploring multi-cloud fleet management |
| `EKS_GKE` | AWS EKS cluster registered with GCP as a GKE Attached Cluster via Fleet + GKE Connect agent via Helm | Engineers exploring multi-cloud fleet management |
| `VMware_Engine` | GCVE private cloud + VMware Engine Network + VPC peering + network policy + firewall rules + Windows jump host + vCenter credential reset | Engineers exploring VMware workload migration to GCP |
| `Container_Migration` | GKE cluster + Compute Engine VMs (PostgreSQL source, Tomcat source, M2C workstation) for Migrate to Containers (M2C) lab | Engineers replatforming VM-based Linux workloads to containers |
| `Migration_Center` | Windows Server VM (MCDCv6 pre-installed) + Debian Linux target VMs + Migration Center service registration + optional AWS asset import | Engineers running Migration Center discovery and TCO assessment labs |
**Supporting directories:**
- `rad-launcher/` — `radlab.py` Python CLI that wraps `tofu`/`terraform` for interactive deployment from a workstation or Cloud Shell.
- `rad-ui/automation/` — Cloud Build YAML files (`cloudbuild_deployment_{create,destroy,purge,update}.yaml`) invoked by the RAD platform UI, plus `check_step_arg_limits.py` and `scripts/` (step logic extracted out of the YAML to stay under Cloud Build's 10,000-character step-arg cap: `apply_infrastructure.sh`, `apply_infrastructure_update.sh`, `prepare_destroy.sh`, `handle_plan_cycle.sh`, and two self-tests). See CLAUDE.md § Deployment Pipelines before editing any of them.
- `scripts/` — standalone helper shell scripts (`gcp-istio-security/`, `gcp-istio-traffic/`, `gcp-cr-mesh/`, `gcp-m2c-vm/`, `gcp-ge-cymbal/`) for lab exercises; not called by any module.
- `SKILLS.md` — detailed implementation guide; read this before making structural changes.
**Standard file layout (using `Istio_GKE` as the canonical example):**
modules/Istio_GKE/ ├── main.tf # Project bootstrap, API enablement, random_id ├── provider-auth.tf # google + google-beta providers with SA impersonation ├── versions.tf # required_providers + required_version ├── variables.tf # UIMeta-annotated inputs (groups 0–4) ├── outputs.tf # deployment_id, project_id, cluster_credentials_cmd, external_ip ├── network.tf # VPC, subnet with secondary ranges, firewall, Cloud Router + NAT ├── gke.tf # GKE cluster, node pool, cluster SA, IAM, kubernetes provider ├── istiosidecar.tf # null_resource installing Istio sidecar mode (conditional) ├── istioambient.tf # null_resource installing Istio ambient mode (conditional) ├── manifests/ # Raw Kubernetes manifests applied as-is └── templates/ # Kubernetes manifest templates rendered by Terraform
**Key conventions:**
- **No secrets in defaults**: Variables like `resource_creator_identity` have default SA values for the RAD platform; never hardcode credentials with sensitive defaults.
- **API enablement**: `google_project_service` must always set `disable_dependent_services = false` and `disable_on_destroy = false`.
- **null_resource pattern**: Create provisioners use `set -eo pipefail`; destroy provisioners use `set +e` with `--ignore-not-found` to be best-effort.
- **UIMeta annotations**: Every `variable` block description ends with `{{UIMeta group=N order=M }}` for the RAD platform UI.
- **File naming**: `.tf` files use `snake_case`; module directories use `PascalCase` / `SCREAMING_SNAKE_CASE`.
- **Copyright headers**: Every `.tf` file begins with the Apache 2.0 license header.
**Available Workflows:**
- `/global` — General repository context (current)
- `/istio` — Istio_GKE module work
- `/bank` — Bank_GKE module work
- `/multicluster` — MC_Bank_GKE module work
- `/attached` — AKS_GKE / EKS_GKE attached-cluster work
- `/troubleshoot` — Diagnostic and troubleshooting work
- `/maintain` — Maintenance and update work
- `/security` — Security audit and hardening
**Action:**
Identify the context of the user's request. If it maps to a specific module or workflow, switch to that workflow. If it is a general question, answer based on the architecture described above. Always consult `SKILLS.md` for implementation details before writing Terraform code.
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.
- yesterday First seen · 559 lines · 13,860 tokens per session scan C ffe0ad7bb58c
rad-modules AGENTS.md is an instructions file published in the GitHub repository techequitycloud/rad-modules (2 stars, last pushed 8d ago), licensed MPL-2.0. It adds 13,860 tokens to every session, about $0.0693 per session on Opus 5. A static security scan graded it C with 2 findings (reaches for credential files, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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.
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).
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.
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).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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.