What is proofrag AGENTS.md?

A guide and command-line tool for evaluating retrieval-augmented generation (RAG), where an AI answers using information retrieved from a document collection. The guide explains how to install and run the proofrag engine with supported AI providers.

Use it to evaluate a RAG system, run the included demo, or install the proofrag command-line engine for project-specific checks.

It provides a repeatable way to examine how well an AI system retrieves information and uses it in answers.

proofrag AGENTS.md — Instructions file by unshDee

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/unshdee/proofrag/agents-md

Clone the repo

git clone --depth 1 https://github.com/unshDee/proofrag

Made for: Codex, OpenCode.

Per session 348 This file is loaded in full into every session.

When invoked 348 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	$0.00348	$0.00348
Opus 5	$0.00174	$0.00174
Sonnet 5	$0.00070	$0.00070
Haiku 4.5	$0.00035	$0.00035

Measured 2d ago against content hash 465ffec24590, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

proofrag AGENTS.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 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.

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.

AGENTS.md · 37 lines

What it actually says

Agents

This repo ships proofrag as a portable Agent Skill: skills/proofrag/SKILL.md. The skill is the interface; the proofrag Python CLI (src/proofrag/) is the engine it drives.

Use it as a skill

Claude Code (plugin):

/plugin marketplace add unshDee/proofrag
/plugin install proofrag@proofrag

Then just ask: "evaluate my RAG" — Claude auto-loads the skill. Or type /proofrag.

Claude Code (manual): copy the skill folder where Claude discovers skills:

cp -r skills/proofrag ~/.claude/skills/        # personal
cp -r skills/proofrag .claude/skills/          # this project only

Codex / other agents (open standard): drop the skill into your agent's skills directory (e.g. .agents/skills/ or your tool's equivalent):

cp -r skills/proofrag .agents/skills/

Install the engine

The skill calls the proofrag CLI. Install it once, or run ad-hoc with uvx:

uv tool install "proofrag[anthropic]"     # or: pipx install "proofrag[anthropic]"
uvx "proofrag[anthropic]" demo            # no install

Set ANTHROPIC_API_KEY (default Haiku) or OPENAI_API_KEY (OPENAI_BASE_URL for local/Ollama). No key needed for proofrag demo.

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.

2d ago First seen · 37 lines · 348 tokens per session scan A 465ffec24590

Subscribe to this mod's changes

proofrag AGENTS.md is an instructions file published in the GitHub repository unshDee/proofrag (2 stars, last pushed 23d ago), licensed MIT. It adds 348 tokens to every session, about $0.0017 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.

Data and AI Evaluation RAG and vectors Testing agent-skills ci claude claude-code codex

AutoRAG copilot-instructions.md

Copilot instructions for Marker-Inc-Korea/AutoRAG, covering autorag - ai-powered rag pipeline optimization, working effectively, prerequisites and system setup, platform dependencies and install java 17 (required).

Marker-Inc-Korea/AutoRAG · 3,237 tokens

GPT-RAG config-python.instructions.md

Instructions for Azure/GPT-RAG, a project described as: Sharing the learning along the way we been gathering to enable Azure OpenAI at enterprise scale in a secure manner. GPT-RAG core is a Retrieval-Augmented Generation pattern running in Azure, using Azure Cognitive Search for retrieval and Azure OpenAI large…

Azure/GPT-RAG · 180 tokens

proofrag AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

Agents

Use it as a skill

Install the engine

What this file has done since we first saw it

Other instructions, from other repositories

AutoRAG copilot-instructions.md

GPT-RAG config-python.instructions.md

GPT-RAG release.instructions.md

eval-view AGENTS.md

rag-code-mcp copilot-instructions.md

evals4j CLAUDE.md