iaf CLAUDE.md

iaf CLAUDE.md is an instructions file for coding agents from dlapiduz/iaf. It costs 1,308 tokens per session, scanned A, original, MIT.

Project instructions for IAF, an application platform designed for AI agents to deploy, run, and manage applications on Kubernetes. Kubernetes is software that schedules and manages containerized applications across machines.

In plain words
What is it for?
Use them when developing IAF features, deciding how workloads or tenants should be handled, and checking that changes meet its build, test, and deployment requirements.
Why use it?
They explain the project's safety and architecture rules, including isolation between users, access control, resource limits, and support for short- and long-running applications.

Instructions file

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/dlapiduz/iaf/claude-md
Clone the repo
git clone --depth 1 https://github.com/dlapiduz/iaf

Wrote 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.

agentmods badge for iaf CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/dlapiduz/iaf/claude-md.svg)](https://agentmods.dev/instructions/dlapiduz/iaf/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/dlapiduz/iaf/claude-md"><img src="https://agentmods.dev/badge/instructions/dlapiduz/iaf/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,308 This file is loaded in full into every session.
When invoked 1,308 The same file — it is already loaded in full.
Security scan A 0 findings. 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 5.1 $0.01308 $0.01308
Opus 5 $0.00654 $0.00654
Sonnet 5 $0.00262 $0.00262
Haiku 4.5 $0.00131 $0.00131

Measured 5d ago against content hash 8037f44e1678, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

iaf CLAUDE.md scanned grade A with 0 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 5d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

IAF - Intelligent Application Fabric

Platform Vision

IAF is an agentic application platform — infrastructure purpose-built for AI agents to deploy, run, and manage applications on Kubernetes.

Key design principles every contributor must understand:

  • Agentic-first: The primary users are AI agents, not humans. APIs, tools, and error messages must be unambiguous, self-describing, and recoverable without human intervention.
  • Short- and long-lived workloads: IAF must support both ephemeral apps (a quick demo, a one-off task) and persistent production services. Lifecycle management, cleanup, and resource limits matter.
  • Multi-tenancy at every layer: Multiple agents (and the humans/organizations behind them) share the same cluster. Namespace isolation, RBAC, and resource quotas are not optional features — they are core invariants.
  • Enterprise-ready security: IAF will be used by small teams and large organizations alike. Security decisions made now are hard to undo. Every feature must be designed with the assumption that a compromised or misbehaving agent must not be able to affect other tenants.
  • Shared resources, clear boundaries: Agents share the cluster, the registry, the ingress controller, and the build system. Any feature that touches shared infrastructure must account for fair use, rate limiting, and blast radius.

Security non-negotiables:

  1. Agent sessions are isolated in their own Kubernetes namespace — never bypass this.
  2. No cross-namespace access to secrets, services, or storage.
  3. All API and MCP endpoints require authentication — never add unauthenticated routes that touch cluster state.
  4. Containers run as non-root. Never relax this in platform-managed workloads.
  5. User-supplied values (app names, env vars, file paths, image references) must be validated and sanitized before use in Kubernetes objects or shell commands.
  6. Least privilege: the platform's own service account should only have the permissions it needs — review RBAC before adding new verbs or resources.

Read the full file on GitHub · 86 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. 5d ago First seen · 86 lines · 1,308 tokens per session scan A 8037f44e1678

Subscribe to this mod's changes

iaf CLAUDE.md is an instructions file published in the GitHub repository dlapiduz/iaf (5 stars, last pushed 3mo ago), licensed MIT. It adds 1,308 tokens to every session, about $0.0065 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.