What is AKF CLAUDE.md?

A set of instructions for AKF, a format that records how trustworthy files are and what checks were performed on them. It tells an AI agent to check this record before using a file and add one after changing it.

Use it when multiple AI agents create or modify files and need to record test results, type checks, reviews, or other verification before handing work on.

It reduces the risk of an agent relying on outdated or unverified work. The recorded evidence helps later agents know whether they need to inspect or test a file again.

AKF CLAUDE.md — Instructions file for Claude Code by HMAKT99: A set of instructions for AKF, a format that…

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/hmakt99/akf/claude-md

Clone the repo

git clone --depth 1 https://github.com/HMAKT99/AKF

Made for: Claude Code.

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/hmakt99/akf/claude-md.svg)](https://agentmods.dev/instructions/hmakt99/akf/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/hmakt99/akf/claude-md"><img src="https://agentmods.dev/badge/instructions/hmakt99/akf/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 828 This file is loaded in full into every session.

When invoked 828 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.

Model	Per session	Once invoked
Fable 5.1	$0.00828	$0.00828
Opus 5	$0.00414	$0.00414
Sonnet 5	$0.00166	$0.00166
Haiku 4.5	$0.00083	$0.00083

Measured 5d ago against content hash 6f0b2fe294e2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

AKF 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 · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AKF — Agent Knowledge Format

AKF is the trust metadata standard for AI-generated content. Stamps are notes agents leave for the next agent: what was done, what was verified, and how much to trust it. A stamp costs ~15 tokens; re-verifying costs thousands.

The protocol

Before building on an existing file — check it

akf check <file>

OK — fresh stamp with verified evidence. Build on it; skip re-verification.
STALE — modified after stamping. Re-verify before trusting.
LOW — stamped but unverified (no test/review evidence). Verify before trusting.
UNSTAMPED — no metadata. Treat as unverified.

After creating or modifying files — stamp what you verified

akf stamp <file> --agent claude-code --evidence "<evidence>"

Evidence is auto-classified and drives trust: "42/42 tests passed" (test_pass), "mypy: 0 errors" (type_check), "reviewed by @user" (human_review), "CI green" (ci_pass). Only claim evidence you actually observed — a stamp that says tests passed when they didn't poisons every agent that trusts it.

Deeper inspection when needed

akf read <file>          # Full trust metadata
akf inspect <file>       # Pretty-printed trust scores

Classification

Use --label confidential for files in */finance/*, */secret/*, */internal/*
Use --label public for README, docs, examples
Default is internal

Available Commands

akf check <file> — One-line trust check (OK/LOW/STALE/UNSTAMPED, exit codes 0/1/2)
akf stamp <file> — Add trust metadata to any file
akf read <file> — Read trust metadata from any file
akf inspect <file> — Pretty-print trust scores
akf embed <file> — Embed metadata into DOCX/PDF/images
akf extract <file> — Extract metadata from any format
akf scan <dir> — Security scan a directory
akf audit <file> — Compliance audit (EU AI Act, SOX, NIST)
akf trust <file> — Compute effective trust scores

Project Structure

python/akf/ — Python SDK source
typescript/src/ — TypeScript SDK source
site/ — Website (Vite + React + Tailwind)
spec/ — Format specification and JSON schema
packages/ — Framework integrations (MCP, LangChain, LlamaIndex, CrewAI)
extensions/ — VS Code, GitHub Action, Office, Google Workspace
skills/ — Agent skill files

Read the full file on GitHub · 76 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.

5d ago First seen · 76 lines · 828 tokens per session scan A 6f0b2fe294e2

Subscribe to this mod's changes

AKF CLAUDE.md is an instructions file published in the GitHub repository HMAKT99/AKF (15 stars, last pushed 1mo ago), licensed MIT. It adds 828 tokens to every session, about $0.0041 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 2026-08-30.

Building agents LLM ai compliance file-format provenance

wasmagent-js CLAUDE.md

Claude Code instructions for WasmAgent/wasmagent-js, covering wasmagent-js — development guide for claude, what this project is (and is not), repository boundaries, this repository owns and other repositories own — do not duplicate here.

WasmAgent/wasmagent-js · 4,124 tokens

agent-3md AGENTS.md

AGENTS.md instructions for CorvidLabs/agent-3md, covering agent-3md, project map, conventions and corvidlabs trust toolchain.

CorvidLabs/agent-3md · 442 tokens

Gloss AGENTS.md

AGENTS.md instructions for RecursiveIntell/Gloss, covering agents.md — gloss closing pass rules, project purpose, source-of-truth ownership, forbidden behavior and required validation.

RecursiveIntell/Gloss · 471 tokens

Gloss CLAUDE.md

Claude Code instructions for RecursiveIntell/Gloss, covering claude.md — gloss project instructions, test integrity contract (non-negotiable), critical spec constraint, scheduling + notebook contracts (non-negotiable) and 1) single-flight llm/gpu gate.

RecursiveIntell/Gloss · 527 tokens

agent-3md CLAUDE.md

Claude Code instructions for CorvidLabs/agent-3md: This repo is governed by the CorvidLabs trust toolchain. See AGENTS.md for the standing rules block and follow it before making changes (run fledge lanes run verify; never bypass the augur, attest, or spec gates).

CorvidLabs/agent-3md · 57 tokens

ai-dial-core CLAUDE.md

Claude Code instructions for epam/ai-dial-core, covering claude.md, build & run, set credentials via environment variables, build (skip tests) and run all tests.

epam/ai-dial-core · 1,214 tokens